Chemical analysis to promote the use of wild from

Magaia, Telma

2015

Link to publication

Citation for published version (APA): Magaia, T. (2015). Chemical analysis to promote the use of wild fruits from Mozambique. Lund University.

Total number of authors: 1

General rights Unless other specific re-use rights are stated the following general rights apply: Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal

Read more about Creative commons licenses: https://creativecommons.org/licenses/ Take down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

LUND UNIVERSITY

PO Box 117 221 00 Lund +46 46-222 00 00 Chemical Analysis to promote the use of Wild Fruits from Mozambique

DEPARTMENT OF FOOD TECHNOLOGY, ENGINEERING & NUTRITION | LUND UNIVERSITY TELMA MAGAIA

Chemical Analysis to promote the use of Wild Fruits from Mozambique

Telma Magaia

2015

Doctoral thesis which, by due permission of the Faculty of Engineering at Lund University, will be publicly defended at the Center for Chemistry and Chemical Engineering on Thursday September 3rd 2015 at 10.15 a.m. in Lecture Hall B.

Faculty opponent Professor Mehari Gebre-Medhin,

Uppsala University Hospital, Sweden Organization Document name

LUND UNIVERSITY DOCTORAL DISSERTATION

Department of Food Technology, Engineering and Date of issue 2015-09-03 Nutrition. P.O. Box 124, SE-22100 Lund, Sweden

Author Telma Magaia Sponsoring organization Swedish International Development Cooperation Agency (SIDA) Title and subtitle Chemical Analysis to promote the use of Wild Fruits from Mozambique Abstract Wild trees have significant cultural and socio-economic value in rural areas of Mozambique. Most of the wild fruits are seasonal and are available mainly in the wet season. Generally they have a short shelf-life and are eaten fresh or after minimal processing; the most common method of preservation is sun-drying. The fruits of Adansonia digitata, kirkii, Salacia kraussii, Sclerocarya birrea, and Vangueria infausta were selected for this study. New data on nutritional components and other characteristics have been obtained. The pH, titratable acidity and the content of soluble solids in the fruit pulps were determined. The organic acids citric, malic and succinic acids were found at various amounts in all pulps. The contents of different mono- and disaccharides were also analysed in the pulps. The protein content was low in all the fruit pulps, as is the case in fruits in general. However, the protein content was high in the kernels of A. digitata and S. birrea, about 30 to 40% on dry matter basis. The total co ntent and relative amounts of the different essential amino acids are with a few exceptions similar to or above that recommended by the WHO for children aged 3 to10 years. The fat content was below 2% in the fruit pulps, while the fat content in A. digitata kernels was almost 40%, and S. birrea kernels about 60%. The kernels of A. digitata and S. birrea are rich in unsaturated fatty acids, and constituted about 68 and 80%, respectively, of the total fat. The A. digitata kernels contained appreciable amounts of essential fatty acids; the amount of linoleic was about 30% and linolenic acid 2%. S. birrea kernels contained about 7% linoleic acid. The fruits contained both insoluble and soluble dietary fibre. The pulp of A. digitata had the highest amount of soluble dietary fibre, around 60% (on dry matter ba sis), while V. infausta pulp had the highest amount of insoluble dietary fibre, around 40%. The kernels contained 3 to 5% phytic acid which may decrease the absorption of minerals. Treatment with phytase reduced the phytic acid cont ent by 20 to 30% after only 15 minutes enzymatic in cubation. Interestingly, almost 50% of the estimated original content of minerals was found in the supernatant after a few minutes’ enzyme incubation. The amount of iron in the pulps ranged from 1 to 9 mg/100 g (on dry matter basis); the highest amount being observed in S. kraussii. The highest iron content, 29 g/100 g DM, was foun d in whole seeds of A. digitata, 29 mg iron/100 g. The A. digitata pulp contained an appreciable amount of calcium, and the kernel also had high content of calcium. In conclusion, data from this study can be used to encourage the increased consumption of these wild f ruits and kernels. In addition, the results of the analysis of the investigated fruits can form the basis for the selection of fruits for wider use, domestication, and processing to extend their shelf-life and for the manufacture of other food products.

Key words: pulps, kernels, consumption, fat, protein, minerals, dietary fibre

Classification system and/or index terms (if any)

Supplementary bibliographical information Language English

ISSN and key title ISBN 978-91-7422-406-1

Recipient’s notes Number of pages 140 Price Security classification

I, the undersigned, being the copyright owner of the abstract of the above-mentioned dissertation, hereby grant to all reference sources permission to publish and disseminate the abstract of the above-mentioned dissertation.

Signature ______Date______Chemical Analysis to promote the use of

Wild Fruits from Mozambique

Telma Magaia

2015

Maputo-Moçambique

Cover photo: Landscape in the Changara district in Mozambique. FrontBack cover: photo: Artistic A basket representation with wild fruits of selectedtwo strands for theof study.the TQ1 polymer with truncated alkyl side-chains and with hydrogen atoms not shown. © Alice Hedström

Back cover: Artistic representation of the PC61BM molecule with hydrogen atoms not shown. © Alice Hedström

Doctoral thesis: “Computational Predictions of Conjugated Polymer Properties for Photovoltaic Applications” © Svante Hedström, 2015 Division of Theoretical Chemistry Lund University P. O. Box 124 S-221 00 Lund Sweden [email protected]

Copyright Telma Magaia* *Full Name: Telma dos Anjos Leví Jamisse Magaia Department of Food Technology, Engineering and Nutrition DepartmentFaculty of ofEngineering, Chemistry, FacultyLund University of Science Lund University ISBN:ISBN 978-91-7422-405-4 978-91-7422-406-1

PrintedPrinted in inSweden Sweden by byMedi Media-Tryck,a-Tryck, Lund Lund University University LundLund 2015 2015

ii

“There are many people who can do big things, but there are very few people who will do the small things.” MotherTeresa

“Há muitas pessoas que podem fazer grandes coisas. Mas há muito poucas pessoas que farão as pequenas coisas.” Madre Teresa de Calcutá

Abstract

Wild fruit trees have significant cultural and socio-economic value in rural areas of Mozambique. Most of the wild fruits are seasonal and are available mainly in the wet season. Generally they have a short shelf-life and are eaten fresh or after minimal processing; the most common method of preservation is sun-drying. The fruits of Adansonia digitata, Landolphia kirkii, Salacia kraussii, Sclerocarya birrea, and Vangueria infausta were selected for this study. These fruits are the most popular, and are consumed in different districts of Mozambique, especially by children, and form part of their normal diet. Besides from eaten fresh, the fruits are mixed with sugar, pressed to make juice, as jam or a kind of dessert, and the pulp and sometimes the kernels are dried and used as flour to make porridge or a sauce. New data on nutritional components and other characteristics have been obtained. The pH, titratable acidity and the content of soluble solids in the fruit pulps were determined as these are of importance for the fruit processing industry. The pulp of all the fruits except S. kraussii had an acidic character. The organic acids citric, malic and succinic acids were found at various amounts in all pulps. The highest amounts of citric acid were found in A. digitata and L. kirkii; above 20 g/kg. Organic acids are responsible for many characteristic fruity tastes and may enhance the absorption of minerals. The contents of different mono- and disaccharides were also analysed in the pulps. The highest amounts of glucose (7.5%) and fructose (5.7%) were found in L. kirkii, while A. digitata contained the highest amount of sucrose (4.3%). The protein content was low in all the fruit pulps, as is the case in fruits in general. However, the protein content was high in the kernels of A. digitata and S. birrea, about 30 to 40% on dry matter basis. For children aged 4 to 8 years, around 80 and 67% of the adequate intake (AI) (defined by the Food and Nutrition Board of the US Institute of Medicine) could be covered by the consumption of 40 g A. digitata kernels or S. birrea kernels, respectively. The total content and relative amounts of the different essential amino acids are with a few exceptions similar to or above that recommended by the WHO for children aged 3 to10 years. The fat content was below 2% in the fruit pulps, while the fat content in A. digitata kernels was almost 40%, and S. birrea kernels about 60%. The kernels of A. digitata and S. birrea are rich in unsaturated fatty acids, and constituted about 68 and 80%, respectively, of the total fat. The A. digitata kernels contained appreciable amounts of essential fatty acids; the amount of linoleic was about 30% and linolenic acid 2%. S. birrea kernels contained about 7% linoleic acid. Estimations showed that 40 g A. digitata kernels can cover the daily intake of omega-6 fatty acids for those aged 4 to 13 years, and about 90% of the requirement for pregnant women. The

i same quantity of A. digitata kernels can provide 60 to 93% of the daily requirement of omega-3 fatty acids for the same groups. The fruits contained both insoluble and soluble dietary fibre. The pulp of A. digitata had the highest amount of soluble dietary fibre, around 60% (on dry matter basis), while V. infausta pulp had the highest amount of insoluble dietary fibre, around 40%. The kernels contained 3 to 5% phytic acid which may decrease the absorption of minerals. Treatment with phytase reduced the phytic acid content by 20 to 30% after only 15 minutes enzymatic incubation. Interestingly, almost 50% of the estimated original content of minerals was found in the supernatant after a few minutes’ enzyme incubation. The amount of iron in the pulps ranged from 1 to 9 mg/100 g (on dry matter basis); the highest amount being observed in S. kraussii. The highest iron content, 29 g/100 g DM, was found in whole seeds of A. digitata, 29 mg iron/100 g, and consumption of 40 g could provide more than 100% of the AI for children and 43% for pregnant women. The A. digitata pulp contained an appreciable amount of calcium, and the kernel also had high content of calcium. Consumption of 100 g of A. digitata pulp would give about 23 to 29% of the AI for children and about 37% for pregnant women. In conclusion, data from this study can be used to encourage the increased consumption of these wild fruits and kernels. In addition, the results of the analysis of the investigated fruits can form the basis for the selection of fruits for wider use, domestication, and processing to extend their shelf-life and for the manufacture of other food products.

ii Popular Scientific Summary

Wild fruits trees species are widely distributed throughout the African countries. Many of these trees species produce fruits, which are used by the local communities to greater or less degrees. The importance of wild fruits in the diet depends to a large extent on the availability of the fruits, since cultivated fruit trees are not particularly common in the dry regions of the countries. The goals of this work were to perform a study on traditional utilization of wild fruits in Mozambique and to generate data on the composition and other characteristics of five wild fruits as a basis for the selection of fruits suitable for processing and increased utilisation and consumption of indigenous fruits. Fruits of Adansonia digitata (n´buvu), Landolphia kirkii (vungwa), Salacia kraussii (psincha), Sclerocarya birrea (canhi), and Vangueria infausta (pfilwa), were selected for this research. When the wild fruits were collected in the different districts, local people were interviewed about the traditional use of the fruits. The fruits are eaten raw, pressed to juice, fermented to alcohol beverages, cooked, or used as flour to make porridge. The seeds and kernels are often used in times of drought when the production of peanuts is low; the seeds can be ground roasted to make a kind of coffee, and the kernels of A. digitata and S. birrea may also be roasted to be eaten as snacks or ground in a bowl, mixed with water, boiled and consumed with local food. In general most people know that fruits are associated with potential health benefits. Researchers has shown that a wide range of wild fruits have the potential to provide rural households with free foods to meet their nutritional and medicinal needs. The disadvantages of the selected fruits are that they grow and ripen during a very short period of the year and that normally their shelf-life is short. Different analytical methods were used to evaluate the nutritional components and other characteristics of some wild fruits. For the fruit processing industry, data on sugar content, pH and moisture are essential characteristics. They play an important part in the perception of fruit quality indicating the possibility for future use of these wild fruits. The sugar content is important for the development of the aroma and taste, and in product development it is important to find a good balance between for example pH and sugars to receive an optimal taste. The highest total sugar content was found in A. digitata and L. kirkii, more than 10g/100 g. Organic acids such as citric and malic acid influence flavor and aroma and are responsible for many characteristic fruity tastes; they were found in all fruits in this study. The parameters above have been reported to influence shelf life, stability and microbiologic safety. Most of the fruit pulps were juicy except of A. digitata, which was very dry; the moisture content was only around 10%. The fat and protein contents in the pulps

iii were low, which is common in fruits. The pulps of A. digitata and V. infausta contained considerable amounts of dietary fibres. The content of the mineral in the kernels such as calcium, iron and zinc were found in considerable levels and the have great contribution for many functions on the human body. Consumption for example of 100 g A. digitata pulp, can provide around 30% of recommended intake of calcium for pregnant women and children, and 23% of the iron for 4 to 13 years old children. Consumption of 40 g A. digitata kernels can provide 20 to 25% of the recommended intake of iron for children and 16 to 20% from S. birrea kernels. In addition the recommended intake of zinc provided 20 to 44% from A. digitata, and 16 to 36 % from S. birrea kernels for children and pregnant women. The fat content in A. digitata and S. birrea kernels was high around 40 to 50%. Interestingly the fatty acid composition in A. digitata can be compared with that in peanuts, while S. birrea can be compared with olive oil. A. digitata kernels contained the essential fatty acids linoleic and linolenic acids and S. birrea kernels contained only the linoleic acids. Consumption of 40 g A. digitata kernels can cover 60 to 90% of the daily intake of essential fatty acids for those aged 4 to 13 years, and pregnant women. Protein is one of the most crucial nutrients to the human body improving varying aspects of health. The protein content in the kernels of A. digitata and S. birrea was around 30 to 40%. In addition, the levels of essential amino acids in the kernels were comparable with the amino acid requirements stated by the World Health Organization (WHO). For children aged 4 to 8 years, 81 and 66% of the recommended intake may be covered by the consumption of 40 g A. digitata kernels and S. birrea kernels, respectively. The contribution to recommended intake for older children and pregnant women is lower, but it may be possible for these groups to increase their intake of kernels, especially if they are eaten as a snack. The results of this study will be vital in efforts to promote use of these wild fruits. The data can be used for the estimation of dietary intakes, and for education of local communities with regard to the nutritional benefits of free sources of food in their environment. The findings indicate that the nutrient contents of the fruits may help to meet the dietary requirements for children and pregnant women in rural areas. In addition they may promote the use of wild fruits and their kernels, and can form the basis for the selection of fruits for further processing, to extend shelf-life and to manufacture new products. Initiatives should be put in place for the selection and domestication of wild fruits trees.

iv Resumo Científico Popular

Árvores de fruteiras silvestres encontram-se amplamente distribuídas por vários Países africanos. Muitas das espécies produzem frutas que são geralmente consumidas pelas comunidades locais. A importância das frutas silvestres na dieta das populações depende em grande medida da sua disponibilidade, tendo em conta que fruteiras cultivadas não são comuns nas regiões secas dos Países. Os objectivos deste trabalho consistiram em realizar um estudo sobre a utilização tradicional das frutas silvestres de Moçambique e produzir dados sobre a composição e demais características de cinco frutas silvestres, como uma base para a selecção de frutas adequadas para o processamento, uso e consumo intensivo. Frutas de Adansonia digitata (n´buvu), Landolphia kirkii (vungwa), Salacia kraussii (psincha), Sclerocarya birrea (canhi), e Vangueria infausta (pfilwa), foram seleccionadas para o estudo. Durante a colheita das frutas em diferentes distritos foram realizadas entrevistas as populações locais sobre a uso tradicional das frutas silvestres. Este tipo de fruta é consumido ao natural, sob forma de sumo, como bebida alcoólica fermentada, utilizado como farinha para a preparação de papas. As sementes e amêndoas são muito usadas em épocas secas quando a produção de amendoim é reduzida. As sementes podem ser moídas para preparar um tipo de café e as amêndoas de A. digitata e S. birrea podem ser torradas para ser consumidas em lanches ou misturadas com água, cozidas e consumidas acompanhando vegetais locais. Em geral a maioria da população conhece a associação das frutas aos possíveis benefícios para a saúde. Investigadores demonstraram que uma grande diversidade de frutas silvestres têm o potencial de fornecer a famílias rurais alimentos para satisfazer as suas necessidades nutricionais e medicinais. A desvantagem destas frutas é o seu crescimento e amadurecimento num período muito curto do ano e o seu reduzido tempo de prateleira. Diferentes métodos analíticos foram usados para a avaliação dos parâmetros nutricionais e demais características das frutas silvestres. Para a indústria de processamento de fruta, dados sobre o teor de açúcares, pH e humidade são parâmetros essenciais. Eles desempenham um papel essencial para a compreensão da qualidade da fruta, indicando possibilidades futuras de utilização. O teor de açúcares é importante para o desenvolvimento do aroma e sabor e no desenvolvimento de produtos é essencial encontrar um bom equilíbrio entre, por exemplo, o pH e os açúcares para obter o melhor sabor. O teor mais alto de açúcares foi encontrado em A. digitata e L. kirkii, 10.3 e 14.4 g/100 g respectivamente. Ácidos orgânicos tais como cítrico e málico influenciam o sabor e o aroma e são responsáveis por muitos sabores característicos das frutas; foram encontrados em todas frutas estudadas.

v Os parâmetros acima descritos têm sido reportados como tendo influencia no tempo de prateleira, estabilidade e segurança microbiológica. As polpas das frutas são em geral suculentas excepto da A. digitata, que é muito seca; o teor de humidade encontrado foi de apenas 10%. O teor de gordura e proteínas nas polpas foi baixo, o que é comum em frutas. As polpas de A. digitata e V. infausta contem quantidades razoáveis de fibra alimentar. Minerais tais como cálcio, ferro e zinco foram encontrados em quantidades consideráveis nas amêndoas e eles possuem grande contributo para muitas funções do organismo humano. A ingestão de 100 g de polpa de A. digitata pode suprir cerca de 30 % do consumo recomendado de cálcio para mulheres grávidas e crianças, e 23% de ferro para crianças de 4 a 13 anos. A ingestão de 40 g de amêndoa de A. digitata pode suprir 20 a 25% do consumo recomendado de ferro para crianças e 16 a 20% se a amêndoa for de S. birrea. Ademais, o consumo recomendado de zinco para crianças e mulheres grávidas pode ser suprido em 20 a 44% por amêndoas de A. digitata, e 16 a 36 % de S. birrea. O teor de gorduras em amêndoas de A. digitata e S. birrea foi elevado, cerca de 40 a 50%. Interessante notar que a composição de ácidos gordos em A. digitata pode ser comparada com a de amendoim, enquanto da S. birrea pode ser comparada a do azeite de oliveira. As amêndoas de A. digitata contêm os ácidos gordos essenciais linoléico e linolênico e de S. birrea apenas o ácido linoléico. A ingestão de 40g de amêndoas de A. digitata pode suprir 60 a 90% do consumo diário recomendado de ácidos gordos essenciais para crianças de 4 a 13 anos e mulheres grávidas. Proteínas são nutrientes cruciais para o organismo humano melhorar diferentes aspectos da saúde. O teor de proteínas em amêndoas de A. digitata e S. birrea foi de 30 to 40%. Além disso, os teores de aminoácidos essenciais nas amêndoas foram comparáveis as exigências de aminoácidos estabelecidas pela Organização Mundial da Saúde (OMS). Para crianças dos 4 aos 8 anos, 81 e 66% do consumo recomendado pode ser suprido pelo consumo de 40 g de amêndoas de A. digitata e S. birrea, respectivamente. A contribuição para o consumo recomendado para crianças com idades acima e mulheres grávidas é baixa, mas seria possível para estes grupos aumentar a sua ingestão, especialmente se as amêndoas forem consumidas regularmente como refeição ligeira. Os resultados deste estudo são essenciais no esforço para a promoção do uso das frutas silvestres. Os dados podem ser usados para estimar o consumo diário e para a educação das comunidades locais sobre os benefícios nutricionais de fontes gratuitas de alimento abundantes no seu meio ambiente. Os resultados indicam que o teor de nutrientes das frutas pode ajudar a suprir as necessidades nutricionais de crianças e mulheres grávidas em áreas rurais. Adicionalmente a promoção da utilização das frutas silvestres e suas amêndoas pode constituir a base para a selecção de frutas para vi processamento adicional, assim como para aumentar o tempo de prateleira e para a preparação de novos produtos. O estudo poderá incentivar iniciativas para a selecção e domesticação de fruteiras silvestres.

vii List of papers

This thesis is based on the results presented in the following papers, referred to in the text by their respective Roman numerals.

I. Proximate Analysis of Five Wild Fruits of Mozambique Telma Magaia, Amália Uamusse, Ingegerd Sjöholm and Kerstin Skog The Scientific World Journal http://dx.doi.org/10.1155/2013/601435 (2013)

II. Dietary Fiber, Organic Acids and Minerals in Selected Wild Edible Fruits of Mozambique Telma Magaia, Amália Uamusse, Ingegerd Sjöholm and Kerstin Skog SpringerPlus 2013, 2:88 (2013)

III. Composition of Amino Acids, Fatty Acids and Dietary Fibre Monomers in Kernels of Adansonia digitata and Sclerocarya birrea Telma Magaia and Kerstin Skog Submitted

IV. Effect of Heating, Lactobacillus Fermentation and Enzyme Treatment on the Content of Phytate in Kernels of Adansonia digitata and Sclerocarya birrea. Telma Magaia, Ingegerd Sjöholm, Estera Dey and Kerstin Skog Submitted

viii The author’s contributions to the papers

I. The author participated in the study design, and carried out the field work, which involved the collection of fruit and information about the traditional use of the fruits. The author carried out the experimental analysis (apart from the sugar analysis), took an active part in data evaluation and wrote the first draft of the manuscript, which was finalized with the contributions of the co-authors.

II. The author participated in the study design. The author carried out the experimental analysis (apart from the mineral analysis), took an active part in data evaluation and wrote the first draft of the manuscript, which was finalized with the contributions of the co-authors.

III. The author participated in the study design. The author carried out the field work, which involved the collection of seeds. The author carried out the experimental analysis (apart from the fatty acids), took an active part in data evaluation and wrote the first draft of the manuscript, which was finalized with the contributions of the co-authors.

IV. The author participated in the study design. The author carried out the experimental analysis (apart from the mineral analysis), the phytic acid analysis was carried out in collaboration with the Department of Food Technology and Nutrition at Chalmers University of Technology in Gothenburg, took an active part in data evaluation and wrote the first draft of the manuscript, which was finalized with the contributions of the co-authors.

Related publication Edible Wild Fruits of Mozambique T. Magaia, J. da Cruz Francisco, A. Uamusse, I. Sjöholm, K. Skog (Proceedings), Acta Horticulturae 948, ISHS, 31, 223-228. 1st International Symposium on Wild Relatives of Subtropical and Temperate Fruit and Nut Crops, 19-23 March, 2011, Davis, California, USA

ix x Contents

Contents 1. Introduction 1 2. Objectives 2 3. Literature Review 3 3.1 Mozambique and wild fruits 3 3.2 General descriptions of fruits 5 3.3 Wild fruits selected for this study 7 3.4 Nutritional data on the selected fruits 14 4. Materials and Methods 25 4.1 Samples 25 4.2 Chemical analysis 25 4.3 Statistical analysis 27 4.4 Interviews regarding the traditional use of wild fruits 27 5. Results and Discussion 28 5.1 Interviews 28 5.2 Chemical analysis 33 5.3 Dietary intake estimations 47 6. Conclusions and Future Outlook 53 Acknowledgements 56 References 58

xi

1. Introduction

The African continent is blessed with a rich tropical flora. A large number of edible wild are distributed throughout Mozambique and local residents are allowed to gather the fruits from the forests. These fruits have significant cultural and socio- economic value, and are important supplemental sources of food in rural areas. The rural population, with a close relationship to the forest, has traditional knowledge on the preparation, consumption and storage of these wild fruits. Some well-known wild fruits in Mozambique are Adansonia digitata (n´buvu), Landolphia kirkii (vungwa), Salacia kraussii (psincha), Sclerocarya birrea (canhi) and Vangueria infausta (pfilwa). These fruits are consumed in different ways, for example, eaten fresh, mixed with sugar, pressed to make juice, as jam or a kind of desert, and the fruit pulp is often dried and used as flour to make porridge. The kernels are often eaten fresh or roasted as snacks. The disadvantage of these fruits is that they grow and ripen during a very short period of the year. They are generally consumed locally, due to their short shelf- life. Transporting the fruits to markets leads to an overabundance, lowering the price and thus income. Since many wild fruits and tubers are underutilized in Mozambique, a programme called “Technology Processing of Natural Resources for Sustainable Development” (TecPro) was initiated in collaboration between the Eduardo Mondlane University in Maputo, Mozambique and Swedish universities. Most wild fruits can be consumed in several ways, but there is little information in the literature regarding the nutritional values of these wild fruits. Thus, there is a need for nutritional data on the wild fruits found in Mozambique in order to increase processing and increase their consump- tion. The work described in this thesis is part of this programme, and the aim was to determine the nutritional components and other characteristics of selected wild fruits commonly consumed in Mozambique.

1 2. Objectives

The overall aims of this work were to study the traditional use of wild fruits in Mozambique, and to obtain data on the chemical composition and other characteristics of five wild fruits as a basis for the selection of fruits suitable for processing. The long-term goal was to promote and increase the use and consumption of indigenous fruits. The fruits studied were: of Adansonia digitata (nbuvu), Landolphia kirkii (vungwa), Salacia kraussii (psincha), Sclerocarya birrea (canhi), and Vangueria infausta (pfilwa). These fruits are popular in Mozambique, where they play an important role in the diet, particularly in rural areas. The specific goals were: • to collect information concerning knowledge on the occurrence of these fruits, and traditional habits regarding consumption and storage, • to obtain data on the contents of dry matter, fat, protein, ash, minerals, insoluble and soluble dietary fibre in the pulps and kernels of the selected fruits, • to determine the pH and titratable acidity in the pulps of the selected fruits, as well as the contents of soluble solids, sugars and organic acids, • to analyse the amino acid composition and the fatty acid composition of the kernels of A. digitata and S birrea, • to use enzymes or lactobacillus species to reduce the phytic acid content in the kernels of A. digitata and S birrea, and • to compare the results with dietary recommendations.

2 3. Literature Review

3.1 Mozambique and wild fruits

Mozambique is located on the eastern coast of Southern Africa on the Indian Ocean (see Figure 1). It has a population of over 26 million, with 68% of the inhabitants living in rural areas (FAO, 2014; FAOSTAT., 2014). The total area of the country is around 800 000 square kilometres, of which 2% is inland water (Mangue and Oreste, 1999). The country stretches in a north-south direction from the Rovuma River to the Ponta d´Ouro. The coastline runs from the tropical to subtropical humid northern and central regions, to the semi-arid/arid subtropical southern region. The climate is characterized by a dry winter season from April to September and a rainy summer season from October to March (FAOSTAT, 2014). The mean annual rainfall ranges from 800 to 1000 mm along the coast and 1200 mm in the central region, to 1000 to 2000 mm in the northern region (Coughlin, 2006; Queface, 2009). An increase in the frequency of droughts, floods and cyclones has had a devastating effect on crops, forest, livestock and fishing in rural and coastal areas (FAO, 2012).

Figure 1. Map of Mozambique showing provinces and cities. The provinces in which the fruits were harvested are indicated by red dots.

3 Many wild fruits are found in Mozambique, which are valuable to the diets and incomes of local communities across sub-Saharan Africa, particularly during times of potential food insecurity. Wild fruits and vegetables play an important role in combating malnutrition and poverty in the African continent. The consumption pattern in Mozambique generally includes three daily meals, such as breakfast, lunch and dinner. However, the number of meals can sometimes be reduced to two or even one when there is a lack of food. Plant foods may provide almost all vitamins and essential minerals, as well as a number of other health-promoting compounds (FAO, 2011). However, nutrient deficiency continues to be a major health problem in developing countries, and has far-reaching consequences on growth, development and health, especially among children, who are most vulnerable to diseases caused by dietary deficiencies (Aphane et al., 2002). Wild fruits can provide an alternative crop when cereal production fails due to lack of rain (FAO, 2010). According to the Food and Agriculture Organization of the United Nations (FAO), “Food security is achieved when everyone has physical, social and economic access to sufficient, safe and nutritious food that meets dietary needs and food preferences for an active and healthy life.” (FAO, 2012). Estimates from 2010 show that about 456 000 people lacked food security in Mozambique, most of them in the Tete province, followed by the provinces of Sofala, Inhambane, Manica, Gaza, Maputo and Nampula. Increased food production and improvements in food security remain high on the list of priorities for Mozambique, whilst simultaneously ensuring sustainable management of natural resources (FAO, 2012). According to a report by the National Directorate of Land and Forests in 2007 (the Ministry of Agriculture in Mozambique), there is great diversity of species of forest products. About 4000 plants have been recognized as being edible plants, most of which are seasonal. The harvesting of these plants and fruits occurs mostly in the rainy season, which coincides with the lean season for agricultural products, thus contributing to the food security of households (Albano and Nhamirre, 2007). Preferences for wild fruits in Mozambique vary according to the region and their cultural value, or the way in which different foods can be prepared from them. In the north, the most commonly found wild fruits are Julbernardia globiflora, Markhamia obtusifolia, Sclerocarya birrea and Tamarindus indica. In the centre of the country, Adansonia digitata, Sclerocarya birrea, Uapaca kirkiana and Ziziphus mucronata, are commonly found, and in the south, Dialium schlechteri, Landolphia kirkii, Sclerocarya birrea, Strychnos madagascariensis, Strychnos spinosa, Trichilia emetica and Vangueria infausta (Mangue and Oreste, 1999).

4 3.2 General descriptions of fruits

Botanically, fruits are the reproductive structures formed by plants, which enclose the seeds and help with their dispersal. Fruits can be categorized as stone fruits or drupes, pip fruits, berries, citrus and aril fruits (Wyk, 2005). A fruit is composed of the skin or peel, the pulp, which is the soft edible part, and seeds that consist of a hard shell and a kernel which is sometimes edible. This is illustrated for Adansonia digitata in Figure 2.

a) b) Figure 2. A whole Adansonia digitata fruit showing the skin a), and in b) the pulp with seeds embedded (A), the pulp (B), whole seeds, i.e. the hard shell + kernel (C), and kernels (D).

Fruits are widespread in nature, and are classified according to climate: (i) temperate, e.g. apples and grapes, (ii) subtropical, e.g. oranges and mandarins, and (iii) tropical, e.g. guava and pineapples (Hoe and Siong, 1999; Hernández et al., 2006; Prasanna et al., 2007). Fruits may contribute significantly to the daily nutrient needs of the individual, however, the amount of each nutrient required by the human body depends on factors such as age, weight, gender, health and level of physical activity (Prasanna et al., 2007). In addition, fruits offer a great variety of aromas, tastes, colour, and texture. Nutritional data for some fruits are presented in Table 1.

5 6 Table 1. Literature data on the nutritional composition of some common fruits (based on wet weight) Climate Energy1 Dry Fat Protein Carbo- Dietary Ash Vitamin C Ca Fe K P Fruit matter2 hydrate fibre (kJ/100 g) (g/100 g) (mg/100 g) Temperate

Apple 251.4 15.4 0.4 0.6 136 0.6 0.3 7.7 9 2.1 - 16 Grape 284.9 18.5 0 0.8 16.3 1.8 0.4 7.6 21 0.5 640 66 Subtropical

Mandarin 205.3 9.3 0.5 0.7 10.4 0.8 0.5 39.6 40 2.1 42 19 Orange 146.7 10.2 0.1 1.5 7.1 0.5 0.1 71 11 0.7 16 3 Tropical Guava 192.7 12.9 0.2 1 10 6.8 0.7 152 33 1.2 12 15 Pineapple 205.3 12.9 0.2 1 10.9 0.4 0.4 28 28 0.3 1490 21 Data from Hoe and Siong, 1999 and Hernández et al., 2006 1 Recalculated from kcal/100 g; 2 Recalculated from moisture %; - = No information given.

6 3.3 Wild fruits selected for this study

The fruits selected for this study are presented below, with photographs of the trees, fruits and seeds, together with general descriptions of the growth locations, the fruit and their consumption and processing.

Adansonia digitata (A. digitata) English name: baobab, local name: n´buvu or malambe

Figure 3. a) The Adansonia digitata tree, and b) the fresh pulp with seeds embedded.

A. digitata is the most widespread of the Adansonia species on the African continent, and is found in the hot, dry savannahs of sub-Saharan Africa. The largest areas of Adansonia are found in Mozambique, , and (PhytoTrade Africa, 2008). In Mozambique, the tree is found in the provinces of Gaza, Inhambane, Maputo, Manica, Nampula, Niassa, Sofala and Tete (De Carvalho, 1968; Da Silva et al., 2004). The trees thrive in woodlands and savannah areas up to 1500 m above sea level and with an annual rainfall of 300 - 500 mm. It is known as the “tree of life”, and has traditionally been a source of food and medicine.

Family: Bombacaceae Tree: very large, 18-25 m tall Harvest time: April to July Shape and size: oval to elliptic, up to 200 mm in length Colour: yellowish grey hairs Pulp: powdery Seed: bean-shaped, numerous, dark-brown containing one small kernel.

7 The pulp has a sweet taste and can be eaten as a snack. It can be mixed with sorghum paste to make the latter more acidic, and the acidic paste is diluted with water to obtain a thick drink. The pulp can be soaked in water or fresh milk adding sugar to form drinks, like juice or full-cream milk, or a kind of yogurt, or frozen to form “ice lollies”. The pulp can also be boiled and mixed with sugar to make porridge (Sidibe and Williams, 2002; Jama et al., 2008; De Caluwé et al., 2009). The pulp serves to enrich homemade soups and sauces, or couscous, and some tribes use it to coagulate milk to make a kind of cheese (Policy Briefing, 2007; Buchmann et al., 2010). The seeds can be crushed and the shells and kernels roasted and finely ground to provide a good substitute for coffee (De Caluwé et al., 2009). The kernels are commonly eaten fresh or roasted as a snack. The kernels can also be ground to flour, which can be added to soups and stews as a thickener, or boiled for a long time, fermented and then dried for later use (Tredgold et al., 1986; Nnam and Obiakor, 2003; Research Council, 2008; De Caluwé et al., 2009; National Buchmann et al., 2010). The kernels can be ground to a paste and added as flavouring to boiled to make a sauce (Jama et al., 2008; Manfredini et al., 2002), or mixed with the pulp to form sweet “milk”, which is boiled and flavoured with honey. In 2008, the EU authorised the use of dried baobab fruit pulp (Adansonia digitata) as a novel food ingredient (Commission Decision, 2008), and in 2009 it was granted GRAS (Generally Regarded as Safe) status in the United States (FDA, 2009). Products containing baobab ingredients are already available in Italy, France, Switzerland, Spain, the UK, Canada and the US (MBB Consulting, 2006). The pulp is considered a good source of vitamin C (approximately six times more than the content in an orange, based on wet weight), it is rich in pectin, making it an attractive “novel food” and beverage thickening agent, minerals (Ca, Mg and P), antioxidants (Jama et al., 2008; Kabore et al., 2011), organic acids (citric, tartaric) and dietary fibre (Maundu et al., 1999; Sidibe and Williams, 2002). The kernels have high contents of protein and fat, and have been found to be a good source of palmitic, oleic and linoleic acid (Osman, 2004).

8 Landolphia kirkii (L. kirkii) English name: wild peach, local name: vungwa.

Figure 4. a) The Landolphia kirkii climbing shrub, and b) the fruit.

Landolphia kirkii is a species of shrubs that can be found in the Democratic Republic of Congo, Malawi, Mozambique, , , Zimbabwe and South Africa. In Mozambique, they can be found in the provinces of Gaza, Inhambane, Manica, Nampula, Niassa and Zambézia (De Carvalho, 1968; Da Silva et al., 2004). Landolphia plants contain milky latex, and L. kirkii was historically the primary source of African rubber. The extracted latex is of good quality with a high content of rubber (www.naturalhub.com, National Research Council, 2008).

Family: Tree: climbing shrub Harvest time: November to February Shape and size; round, 4 - 90 mm diameter Colour: yellow, darkening to yellowish brown skin Pulp: juicy, pleasant and stringy Seeds: numerous seeds embedded in the pulp.

The fruit is commonly eaten fresh together with the seeds, or the seeds are discarded (National Research Council, 2008). Some people use a finger, knife or other utensil to remove the seeds and eat only the pulp (National Research Council, 2008). The fruit can also be used for the production of spirits (www.cde.unibe.ch). No data on its nutrient content were found in the literature.

9 Salacia kraussii (S. kraussii) English name: not found, local name: psincha

a) b) Figure 5. a) The Salacia kraussii shrub and b) the whole fruit and the pulp.

Salacia species are shrubs or small trees that can be found in Mozambique in sandy soils along the coast, in provinces such as Gaza, Inhambane and Maputo (De Carvalho, 1968; Schmidt et al., 2002; Da Silva et al., 2004).

Family: Celastraceae Tree: shrubs, 0.5 m tall Harvest time: December to February Shape and size: round, 50 mm diameter Colour: bright red to orange Pulp: juicy, pleasant and sweet Seeds: 1 - 3 seeds

The fruits are easily accessible to children who are the main consumers, eating the fruits on their way to and from school, or while tending grazing cattle. The fruit is commonly eaten fresh, and the kernels can be dried and eaten (Maundu et al., 1999; De Carvalho, 1968). Several Salacia species have been used in traditional medicine, such as the Ayurvedic system from India (www.ispotnature.org). No data on its nutrient content were found in the literature.

10 Sclerocarya birrea (S. birrea) English name: marula, local name: canhi

Figure 6. a) The Sclerocarya birrea tree, b) the fruit, and c) the seeds embedded in the pulp.

Sclerocarya birrea trees are common in the lower-lying areas of Southern Africa, and are found in all provinces of Mozambique except for the Nampula province (De Carvalho, 1968; Da Silva et al., 2004). In Mozambique, the fruits are not commonly harvested from the tree as they fall to the ground when mature. Some fruits are sweet and others sour, depending on the tree.

Family: Anacardiaceae Tree: large, 9 - 18 m tall Harvest time: January to March Shape and size: plum-shaped, 5 - 35 mm diameter Colour: greenish-yellow Pulp: juicy, tart with a strong, distinctive flavour Seeds: a single large seed containing 2 - 3 hard kernels.

The pulp is commonly eaten fresh as a snack. When children are playing and find a tree with sweet fruits, they suck the juice and discard the seed and peel. The fruits can be made into a delicious jelly, sweet juice, syrups, jam or marmalade, or mixed with cereal or maize porridge (Tredgold et al., 1986; Van Wyk et al., 2002). The fruits can also be used to produce vinegar or fermented to traditional beer or an alcoholic drink called “Ucanhe or Bucanhe” (De Carvalho, 1968; Mangue and Oreste, 1999). Industrially “Amarula”, a cream liqueur, is produced (Wyk, 2005). The seeds are very hard and need careful cracking, but the delicious kernels are highly appreciated and they contain valuable oil. The kernels are widely eaten and very tasty; they are eaten fresh as a snack or ground to a paste and added to vegetables as flavouring. They can be dried or cooked and eaten together with a mixture of dried

11 peanut extracts, red pepper, salt and other spices in the form of a “ bundle” (Glew et al., 1997) The oil is used to treat sliced meat to preserve it as biltong (Tredgold et al., 1986). The pulp is rich in vitamin C, and the amount is higher than in oranges (Wyk, 2005; Borochov-neori et al., 2008; Hillman et al., 2008). The pulp contains small amounts of citric acid and sucrose (Leakey, 1999), and is rich in minerals, the most abundant being calcium, magnesium, potassium and phosphor (Eromosele et al., 1991; Glew et al., 2004). The kernels have a high fat content. They have been reported to be a source of oleic and linoleic acid (Glew et al., 2004), while another report shows the dominating fatty acids to be palmitic, stearic and arachidonic acid (Leakey, 1999). The protein content is around 30% (Leakey, 1999). In addition, the kernels contain compounds with anti-oxidant activity (Glew et al., 2004; Mariod et al., 2004, Bennet, 2006; Kleiman et al., 2008).

12 Vangueria infausta (V. infausta) English name: wild medlar, local name: pfilwa

a) b) c) Figure 7. a) The Vangueria infausta tree, and b) the fruit, and c) the seeds embedded in the pulp.

V. infausta trees occur in abundance in woodlands, valleys and sandy dunes throughout Southern and Eastern Africa, including Madagascar (National Research Council, 2008). In Mozambique, they are found in the provinces of Gaza, Manica, Maputo, Niassa, Sofala, Tete and Zambézia (Da Silva et al., 2004). There are two varieties of the fruit; sweet and sour (www.gutsamba.co.mz).

Family: Rubiaceae Tree: large, 4 - 6 m tall Harvest time: January to May Shape and size: round, up to 20 mm diameter Colour: pale green to brownish-yellow Pulp: juicy, soft with a strong, distinctive flavour Seeds: 2 - 3 large seed hard kernels.

The fruits are mostly eaten fresh as a snack or soaked in water and mixed with sugar. The pulp can be boiled to make a kind of porridge, or mixed with milk for deserts, or mixed with a little water and sugar to produce an acceptable substitute for apple sauce. The pulp is also fermented to a strong alcoholic drink or brandy (Tredgold et al., 1986; Maundu et al., 1999; Motlhanka et al., 2008,). Industrially, the fruit is processed to make liquors, jams and marmalades (www.gutsamba.co.mz). The whole fruit can be sun-dried and stored for use in times of food scarcity. The most common method of preservation is fermentation to produce alcoholic beverages (Boletim IIAM, 2008; www.cde.unibe.ch; Ibnouf, 2012). The pulp has high contents of dietary fibre, potassium and vitamin C (Amarteifio and Mosase, 2006).

13 3.4 Nutritional data on the selected fruits

Many wild fruits and nuts are good sources of fat, protein, carbohydrates, dietary fibre, vitamins and minerals, but literature data are scarce. Nutritional data for the wild fruits included in this study, are compiled in Tables 2 to 6. The literature data regarding energy, dry matter, fat, protein, carbohydrates, dietary fibre, ash and vitamin C are given in Table 2a for the pulps and Table 2b for the seeds and kernels. Lack of energy in the diet is a major problem in many countries. Energy is supplied by carbohydrates, proteins, fats and sometimes alcohol in the diet (Kennedy et al., 2003). Fat is an important component in the diet as it provides the body with energy in a concentrated form. In addition, dietary fats provide essential fatty acids and fat-soluble vitamins (Insel, et al., 2011). Protein is another important component in the diet. Proteins are involved in the growth and repair of tissues, as enzymes in the metabolism, as hormones, transporters, antibodies, and in many other ways (Whitney and Rolfes, 1996). An inadequate supply of protein is considered to be responsible for malnutrition among people living in developing countries. Dietary fibre alters the water content, viscosity and microbial mass of the intestinal contents (Elleuch et al., 2011). Dietary fibre reduces the risk of heart disease, improves glucose tolerance, by delaying the transport of carbohydrates into the small intestine (Anderson et al., 1994; Rodríguez et al., 2006). Insoluble fibre increases faecal bulk and decreases intestinal transit time (Hamilton et al., 1992; Al-Farsi et al., 2005). Soluble fibre increases viscosity, and reduces the glycaemic response and plasma cholesterol (Abdul-Hamid and Luan, 2000). Although dietary fibre provides many health benefits, it may reduce mineral absorption due to the ability of fibre to bind cations (López and Martos, 2004). Vitamin C has many functions in the body. It is involved in collagen synthesis and amino acid metabolism, it strengthens resistance to infection, and helps in the absorption of iron (Whitney and Rolfes, 1996).

14 Table 2a. Literature data on the nutrients in the wild fruit pulps included in this study Fruit Energy Dry Fat Protein Carbo- Dietary Ash Vitamin C References matter hydrate fibre Based on (kJ/100 g) (g/100 g) (mg/100g) A. digitata FW - 93.0 0.3 3.2 75.9 8.7 5.1 213 Wehmeyer, 1996 DW 1480 93.3 0.2 2.6 79.3 5.7 5.3 300 Nour et al., 1980 DW - 86.8 4.3 3.1 79.4 8.3 5.0 - Saka and Msonthi, 1994 DW 849 89.5 0.4 2.2 70.0 11.2 5.7 - Lockett et al., 2000 DW 1341 95.3 0.7 2.5 - 45.1 5.1 - Murray et al., 2001 FW - 89.6 0.3 3.2 - 5.4 4.5 - Osman, 2004 FW - 86.0 - 1.3 - 16.2a 4.6 141.3 Amateifio & Mosasee, 2006 FW 488-631 86.3-86.7 0.4-0.5 2.03-2.04 78.3-78.4 51.4-52.2 5.5 74.1-76.2 Phyto Trade Africa, 2008 FW - 86.0-89.0 0.5-0.7 2.0-3.2 78.3-78.9 46-54 5.0-7.0 74.0-168 FDA, 2009 FW - 88.8 0.4 3.5 74.3 6.1 4.5 - Oyeleke et al., 2012 S. birrea FW - 8.3 0.1 0.5 7.0 0.5 0.5 67.9 Wehmeyer, 1966 DW - 7.0 - 3.6 49.9 37.7 6.8 Murray et al., 2001 FW - 11.6 - 3.7 - 16.3 a 4.9 128.3 Amarteifio & Mosase, 2006 FW 17.92 12.7 3.4 3.6 84.1 6.4 3.5 190.0 Wairagu et al., 2013 V. infausta DW 1445 26.5 2.6 5.7 78.1 10.2 3.4 - Saka & Msonthi, 1994 FW - 23.5 - 3.0 - 39.4a - 67.7 Amarteifio & Mosase, 2006 FW - 93.7 1.2 3.0 77.1 10.3 3.4 11.5 Emmanuel et al., 2011 FW - 19.3-80.9 - 5.4-5.8 - - - - Mothapo et al., 2014 FW = fresh weight; DW = dry weight; - = no information given; a Dietary fibre analysed as acid detergent fibre. 15

15 16 Table 2b. Literature data on the nutrients in the wild fruit seeds and kernels included in this study Fruit part Energy Dry Fat Protein Dietary Carbo- Ash Vitamin References matter fibre hydrate C Based on (kJ/100 g) (g/100 g) (mg/100g) Seeds of A. digitata DW - 93.9 16.1 14.4 56.8 - 6.1 - Proll et al., 1998 DW - 8.2 11.6 15.1 17.8 49.7 5.8 - Lockett et al., 2000 FW 1525 95.7 12.2 18.4 45.1 16.1 3.8 - Osman, 2004 DW 1871 95.3 12.7 21.7 52.5 6.7 5.0 6.7 Nkafamiya et al., 2007 DW 2169 95.0 38.0 38.2 6.4 10.0 7.5 - Mitchikpe et al., 2008 FW 1820 95.8 6.9 48.3 21.9 4.8 3.8 - Adubiaro et al., 2011 FW - 88.8 3.9 2.5 70.7 9.9 5.4 57.3 Emmanuel et al., 2011 FW - 96.5 13.4 19.5 44.6 15.6 3.1 - Oyeleke et al., 2012 Kernels of A. digitata DW 1898 95.2 29.3 36.3 - 14.1 9.1 - Murray et al., 2001 FW 1966 93.6 23.5 26.7 37.9 - 5.5 - Igboeli et al., 1997 DW - - 18.9 25.5 48.1 - 7.6 - Nnam & Obiakor, 2003 FW - 91.9 34.1 32.7 30.0 - 5.0 - Obizoba & Amaechi, 1993 Kernels of S. birrea FW - 96.0 57.0 30.9 1.5 2.4 4.2 - Wehmeyer, 1966 DW - 97.5 11.0 36.7 17.2 3.4 11.7 - Ogbobe, 1992 DW - 93.9 42.0 30.8 17.3 3.8 - Eromosele & Eromosele 1993 DW 665.7 89.5 58.9 31.0 3.0 2.5 4.7 - Muhammad et al., 2011 FW = fresh weight; DW = dry weight; - = no information given.

16 Tables 3a and b list the contents of some minerals in the fruits studied. No data could be found for L. kirkii and S. kraussii. The minerals included in the Tables are of great importance in the diet. Some minerals or macro-elements are required in amounts greater than 100 mg per day (Insel et al., 2011; Imelouane et al., 2011); these include calcium, phosphorus, magnesium, potassium and sodium. Essential trace elements such as zinc, iron, iodine and selenium are normally required in amounts of less than 100 mg per day (Imelouane et al., 2011). Calcium, together with potassium, phosphor and magnesium, is important for the growth and maintenance of bone and teeth (Insel et al., 2011). Calcium is also involved in muscle contraction and relaxation, nerve function and the immune defence (Whitney and Rolfes, 1996). Iron is necessary for the transport of oxygen in the blood to the cells, for immune and nerve function, and it is a cofactor in numerous enzyme reactions (Insel et al., 2011). Magnesium is involved in more than 300 biochemical reactions in the body, for example, in energy metabolism, protein synthesis, muscle contraction and bone mineralization (Insel et al., 2011). Zinc increases the affinity of haemoglobin for oxygen, participates in the transport of vitamin A, plays a role in taste perception, and interacts with a number of hormones. In addition, the body needs zinc to grow and develop properly during pregnancy, infancy and childhood (Brown et al., 1999; Insel et al., 2011).

17 17 Table 3a. Literature data for some minerals in A. digitata, S. birrea and V. infausta pulp Fruit Ca Fe K P Mg Na Zn References Based on (mg/100 g) A. digitata FW 387 2.2 737 41.9 180 Trace - Wehmeyer, 1966 DW 655 8.6 - 50.8 - - - Nour et al., 1980 DW 60 4.4 - 5.0 209 - 2.4 Eromosele et al., 1991 DW 116 5.8 2836 45 209 18.8 - Saka & Masonthi, 1994 DW 341 1.7 - 73.3 209 5.46 1.0 Glew et al., 1997 DW 216 2.9 726 452 300 0.79 3.2 Sena et al., 1998 DW 211 4.2 - 49.8 123 - 0.5 Lockett et al., 2000 FW 295 9.3 1240 - 90 27.9 1.8 Osman, 2004 FW 128 0.1 1866 50 121 13.3 0.1 Amarteifio, Mosase, 2006 330- 9.02- 2010- 61.1- 143- 0.82- PhytoTrade Africa, 2008 FW - 338 9.13 2030 62.1 146 1.21 257- 4.0- 2010- 56- 126- 0.7- FDA, 2009 DW - 370 9.1 2390 73.3 179 3.1 FW 78.2 5.9 1410 105 69.1 35.1 ND Oyeleke et al., 2012 FW = fresh weight; DW = dry weight; - = no information given; ND = not detected (< 10 mg/g dry weight).

18 Table 3a. (continued) Literature data for some minerals in A. digitata, S. birrea and V. infausta pulp

Ca Fe K P Mg Na Zn Based References on (mg/100 g)

S. birrea FW 6.2 0.1 54.8 18.7 10.5 Trace - Wehmeyer, 1966 DW 36.2 1.1 - 18 31.9 - 0.34 Eromosele et al., 1991 DW 481 2.5 - 264 310 1.52 ND Glew et al., 1997

FW 94 0.07 2183 69 158 13 0.13 Amarteifio, Mosase, 2006 FW 51.7 8.8 44.5 0.2 24.5 14.9 3.0 Hassan et al., 2010 FW 688 2.79 3220 146 158 - 2.3 Wairagu et al., 2013

V. infausta

DW 13.2 28.3 181 82.3 182 24.5 - Saka & 1 Masonthi, 1994 FW 124 0.09 1683 128 99 13.7 0.02 Amarteifio & Mosase, 2006 DW 186 24.4 747 86.9 18.6 161 7.1 Emmanuel et al., 2011 FW 0.2 20.1-21.6 2.2 0.2-0.3 - - - Mothapo et al., 2014 FW = fresh weight; DW = dry weight; - = no information given; ND = not detected (< 10 mg/g dry weight).

19 Table 3b. Literature data for some minerals in the seeds and kernels of A. digitata and S. birrea Based Ca Fe K P Mg Na Zn References on (mg/100 g) Seeds of A. digitata

DW 395 1.8 - 614 352 1.94 2.6 Glew et al., 1997

DW 264 4.4 - 678 278 - 4.3 Lockett et al., 2000

FW 410 6.4 910 - 270 28.3 5.2 Osman, 2004

DW 58.9 6.4 280 6.0 - 6.07 3.6 Nkafamyia et al., 2007 DW 348 10.3 1029 1767 782 - 7.9 Mitchikpe et a., 2008 FW 242 22.0 536 480 352 8.4 3.4 Adubiaro et al., 2011 DW 677 21.7 1013 53.6 66.5 127 4.0 Emmanuel et al., 2011 FW 521 10.1 875 125 315 40.7 - Oyeleke et al., 2012 Kernels of A. digitata FW 2.0 1.3 - 0.2 - - 1.9 Obizoba & Amaechi, 1993 DW 0.5 0.6 0.6 326 - - 1.29 Nnam & Obiakor, 2003 Kernels of S. birrea FW 106 0.42 677 836 467 338 - Wehmeyer, 1966

DW 156 2.8 - 212 193 1.2 2.65 Glew et al., 1997

DW 403 27.5 366 2.9 206 4.8 3.3 Muhammad et al., 2011 DW 154 2.8 364 1040 421 4.3 6.24 Glew et al., 2004

FW 448.9 4.5 486.6 160.7 253.7 - 4.5 Wairagu et al., 2013 FW = fresh weight; DW = dry weight; - = no information given.

20 Many seeds and kernels contain phytic acid, which may have a negative influence on the uptake of minerals in the diet. Phytic acid, also known as inositol hexaphosphate, has six PO4 groups, and is a strong chelator of divalent ions, especially zinc and iron, and to a lesser extent, also calcium and magnesium (Greiner and Konietzny, 1998). Phytic acid is the principal storage form of phosphorus in many plant tissues, especially bran and seeds (Anastasio et al., 2010). Table 4 gives the content of phytic acid reported in the literature for the fruits studied.

Table 4. Literature data on the phytic acid content of the seeds of A. digitata and kernels of S. birrea Phytic acid (%) Based on Reference A. digitata 6.66 and 7.13 - Adubiaro et al., 2011

4.90a DW Mitchikpe et a., 2008

1.75 and 0.62 - Saulawa et al., 2014

1.40a - Proll et al., 1998

1.20 - Ezeagu, 2005

0.20 - Nkafamyia et al., 2007

0.073a FW Osman, 2004

0.18 and 0.16b - Nnam & Obiakor, 2003

S. birrea

0.423a DW Muhammad et al., 2011 a= recalculated from other units; b = units not stated. FW = fresh weight; DW= dry weight; - = not given.

21 Not only the protein content, but also the amino acid profile is important. The essential amino acids cannot be synthesized in the body and must therefore be provided through the diet. Amino acids serve as precursors for proteins, nucleic acids, hormones and other important molecules (Insel et al., 2011). The contents and ratios of essential amino acids are important in the estimation of protein quality (WHO, 2007). Table 5 gives the contents of amino acids in the seeds and kernels of A .digitata and S. birrea.

Table 5. The content of amino acids in the seeds and kernels of A. digitata and S. birrea; essential amino acids are indicated by an asterisk Amino Glew et Osman, Ezeagu, Glew et Glew et Muhammad et acid al., 1997 2004 2005 al., 1997 al., 2004 al., 2011 A. digitata, whole seeds S. birrea, kernels (mg/g DM) (g/100 g of protein) (mg/g (g/100 g of protein) DM) Cysteine* 3.6 1.5 1.9 1.9 2.5 2.4

Histidine* 5.1 2.2 2.0 1.2 2.5 2.6

Isoleucine* 8.3 3.6 3.5 2.5 3.3 3.8

Leucine* 14.0 7.0 6.5 3.8 4.8 7.4

Lysine* 11.2 5.0 3.7 1.3 2.0 5.2

Methionine* 2.3 1.0 1.3 0.7 1.6 1.7

Phenylalanine* 10.3 4.0 4.5 2.4 3.6 3.1

Threonine* 7.0 3.8 2.9 1.8 2.4 4.7

Valine* 11.6 5.9 5.0 3.0 3.9 3.9

Alanine 2.5 7.1 - 2.5 2.5 6.2

Arginine 2.2 8.0 - 6.8 14.4 15.0

Aspartic acid 21.2 10.3 - 5.2 12.7 13.4

Glycine 10.4 8.6 - 2.7 3.6 3.5

Glutamic acid 48.9 23.7 - 13.1 29.9 28.1

Proline 9.6 6.9 - 2.6 2.8 4.6

Serine 11.4 6.1 - 2.6 4.1 4.3

Tyrosine 5.6 1.5 2.7 1.5 2.0 2.6 *Essential amino acids, DM= dry matter; - = no information given.

22 The quality of the fat in the diet is also important, especially the content of unsaturated fatty acids. The contents of fatty acids in the kernels of A. digitata and S. birrea are given in Tables 6a and b. The essential fatty acids linoleic acid and linolenic acid are indicated by an asterisk. They are of importance, for example, for the maintenance of the immune system and cell membranes, and for the function of the brain and skin (Uauy et al., 2001). They may also reduce the risk of heart disease (Food Nutrition Board, 2002). The consumption of monounsaturated fatty acids has been associated with decreased levels of low-density lipoprotein cholesterol, and possibly with increased high-density lipoprotein cholesterol (Coultate, 2009).

Table 6a. Literature data on the fatty acids in the whole seeds of A. digitata

Fatty Glew Eteshola & Ezeagu Osman, Ezeagu, Zimba

1 et al., 1997 Oroedu, 1996 et al., 1998 2004 2005 et al., 2005 acid

(mg/g DM) (g/100 g of fat)

C14:0 Trace 38.4 - 0.2 0.3 31.6

C16:0 1.4 19.7 15.5 24.2 22.1 -

C16:1 0.02 - 0.2 - 0.3 3.5

C18:0 0.2 3.2 3.1 4.6 4.0 34.5

C18:1 2.1 22.4 24.7 35.8 35.0 27.8

C18:2 1.4 16.2 19.1 30.7 26.1 -

C18:3* 0.02 - 1.6 1.0 2.0 -

C20:0 Trace - 0.7 1.3 0.9 -

C20:1 - - - 0.9 0.2 -

C22:0 - - 0.4 - 0.4 -

C24:0 - - 0.3 0.2 - - 1 No. of carbon atoms and double bonds, for systematic names see Table 13. *Essential fatty acids, DM= dry matter; - = no information given.

23 Table 6b. Literature data for the fatty acids in kernels of S. birrea. For references A-H, see footnote1 below

Fatty acid2 A B C D E F G H

(mg/g DM) (g/100 g of fat)

C14:0 0.03 2.1 0.1 0.3 - - 0.1 -

C16:0 2.1 22.5 15.6 14.2 10.7 8.7 5.7 12.8

C16:1 0.02 - 0.2 0.2 - - 1.5 -

C18:0 1.1 50.7 11.1 8.8 7.2 9.5 6.0 7.2

C18:1 6.3 4.1 63.2 67.3 72.0 78.5 71.8 73.6

C18:2* 0.5 - 5.2 5.9 8.8 3.4 0.6 6.1

C18:3* 0.03 - - 0.1 0.6 - 0.1 0.3

C20:0 0.06 8.5 1.3 0.9 - - 3.9 -

C20:1 - 0.1 0.5 0.4 1.3 - 2.2 -

C22:0 - 5.1 0.4 0.2 0.1 - 6.6 -

C24:0 - 4.1 0.3 0.3 - - 1.0 - 1References: A: Glew et al., 1997; B: Ogbobe, 1992; C: Glew et al., 2004; D: Mariod et al., 2004; E: Zharare & Dhlamine, 2004; F: Zimba et al., 2005; G: Kleiman et al., 2008; H: Robison et al., 2012.

2 No. of carbon atoms and double bonds, for systematic names see Table 13 *Essential fatty acids, DM= dry matter; - = no information given.

24 4. Materials and Methods

4.1 Samples

The five wild fruit varieties included in this study were collected in different regions of Mozambique. Ripe fruits were collected in 2008 and 2009, except for the fruits from S. birrea, which were collected only in 2009. A. digitata fruits grown in the Tete province were bought at a local market in Maputo, and some fruits were collected from family orchards in the Vilanculos district. L. kirkii, S. kraussii and V. infausta fruits were collected from orchards in the Marracuene and Manhiça districts of Maputo province. S. birrea fruits were obtained from a garden in Maputo city, and seeds, dried for 9-10 months, from a family orchard in Manhiça. For a more detailed description, see Paper I. In addition, seeds from A. digitata and S. birrea were collected in 2013, in Tete and Maputo (see Paper III). Unblemished fruits were selected and washed, the skin and seeds were removed and the pulp was homogenized in a blender. The seeds from A. digitata and S. birrea were crushed, the shells were removed and the kernels inside were collected. Samples for the determination of pH, soluble solids (°Bx) and titratable acidity were kept at room temperature and the analyses were performed the day after collecting the fruit. Samples for other analyses were vacuum packed in plastic bags, frozen and stored at -18°C in a freezer. The kernels were milled before analysis.

4.2 Chemical analysis

The contents of dry matter, fat, protein, dietary fibre, ash and minerals were determined in the pulp and kernels. The pH and titratable acidity and the contents of soluble solids, sugars and organic acids were determined in the pulp. Amino acids, fatty acids and phytic acid were analysed in the kernels of A. digitata and S. birrea. Table 7 gives an overview of the various methods and equipment used for analysis. For a more detailed description of the methods, instruments and chemicals, see Papers I - IV. Each parameter was determined at least in duplicate.

25 Table 7. Analytical methods and equipment

Parameter Method/Equipment Reference Amino acids Ion-exchange chromatography See Paper III with post-column derivatization using ninhydrin Ash Muffle furnace, 24 h at 550°C AOAC, 2000; 940.26

Dry matter Vacuum oven, 24 h at 70°C AOAC, 2000; 926.12 Oven, 18 h at 105°C until AOAC, 1980; 22.021b constant weight Dietary fibre and Enzymatic gravimetric method Asp et al., 1983; Theander et al., their monomers and gas chromatography 1995; Haskå et al.,2008

Fat Soxhlet extraction, petroleum AOAC, 1996; 920.39 ether Fatty acids Gas chromatography Analysed by an authorised laboratory in Sweden Minerals Inductively coupled plasma- Analysed by the Department of atomic emission spectrometry Ecology, Lund University Organic acids Ion-exchange chromatography See Paper II Phytic acid High-performance ion- Carlsson et al., 2001 chromatography pH pH meter

Protein Flash elemental analyser AOAC, 1980; 7.016

Soluble solids Digital refractometer AOAC, 2000; 932.12

Sugars High-performance anion-exchange Analysed by an authorised chromatography laboratory in Sweden

Titratable acidity Titration AOAC, 2000; 942.15

26 4.3 Statistical analysis

Statistical analyses were performed using SPSS (version 13). Significant differences were evaluated with one-way analysis of variance (ANOVA) followed by Tukey’s multiple comparison test (Papers I and II). Microsoft Excel® was used for statistical evaluation. Student’s t-test was performed to determine significant differences. (Papers III and IV). A value of p<0.05 was considered to indicate statistical significance.

4.4 Interviews regarding the traditional use of wild fruits

When the wild fruits were collected in the different districts, local people were interviewed about the traditional use of the fruits selected for this study. The local authority was contacted for permission for the field work. During the first period of field work, in 2008 and 2009, 3 to 10 people were interviewed about each fruit, both women and men of different ages, all of whom had grown up in the rural area. The aim of the interviews was to obtain information on knowledge concerning the occurrence of these wild fruit trees and traditional habits regarding consumption, preparation and storage of the wild fruits. The second period of field work, in 2013, was organized so as to obtain information concerning the use, consumption and preparation of the seeds and kernels. Meetings were organized with 15 to 17 people of different ages (18 to 70), including the local guide who translated from the local language to Portuguese. The main interview questions were the following: (i) Where are the fruits collected? (ii) What is the importance of the fruits? (iii) Who are the major consumers? (iv) Apart from consumption, what is the contribution of the fruits in the household economy? (v) Which parts of the fruits are eaten? (vi) What kind of preparation is usually done? (vii) How are the fruits, seeds and kernels preserved for storage?

27 27 5. Results and Discussion

5.1 Interviews

The interviews revealed that the majority of the fruits came from the forest, or in the area surrounding their house. In periods of food scarcity, the leaves, fruits and seeds are used to provide food for everyone, as well as for medical applications. The fruits are commonly consumed by people of all ages as snacks while walking to, and working on, their grassland. However, the children are the main consumers, since they are more free to go into the forest to collect the fruits while tending grazing cattle or on their way to and from school. The wild fruits harvested serve as a source of income for many families, depending on the amount of fruit that each family can gather. The fruits are usually sold along the roadside (Figure 9), or transported for sale early in the morning at informal markets, for example “Bobole” in Marracuene and “Kwatsena” in the Tete province, to which people come to buy fruits for later sale in the markets in the cities.

a) b) Figure 9. a) A. digitata fruits along the roadside. b) Other common fruits along the roadside.

The pulps of all the studied fruits are consumed, and the seeds and kernels from A. digitata are consumed in some areas. The kernels of S. birrea are sometimes, but not often consumed.

28 Fruit pulps In general, the pulps are eaten fresh as snacks, or squeezed to make juice. They are also fermented to produce local alcoholic drinks that are often drunk at social and traditional events. The pulps of the fruits are also mixed with water to make a juice. The juice prepared from A. digitata pulp is often consumed in “tropical nhungwé” restaurants in the Tete province. In Vilanculos, the juice is filtered and sugar is added, and then packed in small plastic bags, frozen and sold in informal markets, where it is served as a refreshment consumed by both children and adults. Pulps from fruits such as A. digitata and V. infausta are mixed with cow’s milk to make yogurt or a dessert, depending on the amount of liquid added. The squeezing process is not efficient enough to obtain a high juice yield, as part of the flesh is attached to the central pit and skin. Furthermore, it is difficult to obtain consistently ripe and undamaged fruit, and using fruits at different degrees of ripeness could result in a final juice being too sour to be drinkable. The pulps from A. digitata and V. infausta fruit are also boiled to make a porridge given to children before they go to school. This is considered a very nutritious food, aiding the development of infants and children. A. digitata pulp can be processed by adding a little water to obtain a kind of sauce, which is added to dried meat and then boiled. It can be used to make a chili sauce by boiling the pulp with a little water, and mixing it with fried garlic and chili powder, and cooking for a while. The mixture is left in the sun for a few days and can be eaten with any kind of food. The preservation of the fruit is problematic, especially for the L. kirkii, S. kraussii and S. birrea fruits, since they have low dry matter contents and a short shelf-life. V. infausta pulp or fruit can be sun-dried and stored for up to a year for later use in periods of hunger. Whole fruits of A. digitata can be preserved in cool places or barns. Barns are preferred for storage because this avoids contact with the sand, and the fruits are not infested by insects causing poor quality of the fruit pulp. A. digitata can also be stored in special containers or in bags at room temperature for more than a year, without showing any indication of damage by insects. The traditional alcoholic drink made from S. birrea can be stored in sealed clay pots and buried, allowing storage for two years. This was a traditional habit of people in southern Mozambique, particularly in the provinces of Maputo, Gaza and Inhambane, where the majority of the men work for the mining companies in neighbouring South Africa, and come home for a holiday every 18 months.

29 Seeds and kernels The use and benefits of the seeds and kernels of A. digitata and S. birrea were also reported by the elders. The kernels can be eaten raw or mixed with a little salt, roasted and consumed as a snack. The seeds and kernels are often used in times of drought when the production of peanuts is low. The seeds of A. digitata are generally cracked using two stones or a sharp axe, and then beaten to open the seed and separate the kernel. The seeds of S. birrea can be cracked using two stones, or a hammer and a piece of wood from the tree. The kernels are removed from the shell using a stick as a kind of needle to poke them out (Figure 10).

b) c)

Figure 10. a) The cracking of S. birrea seeds, b) removing the kernels from the shell, and c) the kernels.

After squeezing S. birrea fruits for the preparation of beverages, the seeds can be left on the ground to dry for several months for later use. Roasted kernels of S. birrea can be stored in glass jars sealed with a stopper for more than two years, without any change in the taste or aroma. S. birrea kernels have a nice nutty flavour and pleasant smell, providing a condiment with a good taste when mixed with indigenous food plants and boiled for consumption. Due to the difficult and laborious task of obtaining these kernels, the women prepare them as special gifts for their husbands, as a gesture of their affection. In meal preparation, flour from S. birrea kernels is some- times mixed with peanut flour to give the latter more flavour. The locals mentioned that the amount of S. birrea kernel flour needed to give a good sauce is generally smaller than the amount of peanut flour required. S. birrea kernels are also mixed with a little salt, roasted and consumed as a snack. The whole seeds of A. digitata can be ground, mixed with corn or roasted to make a kind of coffee. The kernels may also be ground in a bowl, mixed with water in a pan with local plant food, cooked for a while, and then consumed with corn flour or

30 sorghum. A. digitata seeds are often left on the ground after consumption of the pulp, and it is not common to store them. It is not common to extract oil from either seed of the kernels, although they have a high fat content (Glew et al., 1997). The major problem related with oil extraction is a lack of sophistication in the technique to be used (MMB Consulting, 2006).

31

32 5.2 Chemical analysis

The results of the determinations of dry matter (DM) content, fat, protein, dietary fibre monomers and ash, expressed on DM basis, are given in Table 8 for the pulps and kernels investigated. For comparison with literature data, see Table 2. Generally, no significant differences were seen between the results of the analysis of the fruits collected in the two years (2008 and 2009) at different locations. The small variations in the data could be due to differences in climate, soil and weather conditions. The protein values in the kernels collected in 2013 is lower than those found in the first studies. When the amino acid composition was analysed it became obvious that the kernels had high contents of glutamic acid (with additional nitrogen groups) and thus another protein conversion factor (5.75) was used; in the first studies the common factor 6.25 was used. Also the fat content in the kernels from 2013 was somewhat lower, which may be explained by a different analytical method. The determination of the DM content the day after collecting the fruits showed that the pulp of A. digitata had a very high DM content, almost 90 g/100 g. This is in accordance with results from other studies (Table 2). The result for V. infausta pulp was somewhat higher than the literature data. No data could be found in the literature for the other three fruits. For comparison, the DM content of apples and grapes varies between 15 – 19%, See Table 1. High DM contents in fruits indicate long shelf-life (Effiong and Udo, 2010). In general, fruit pulps have low contents of fat and protein (Insel et al., 2011), see also Table 1. The fat contents in the pulps were below 2 g/100 g DM, in agreement with previous reports (Table 2). The protein contents were below 5 g/100 g DM. In other studies, the protein content has been reported to range from 1 to 6 g/100 g (Table 2). The DM contents of the kernels of A. digitata and S. birrea were high, over 90 g/100 g. These results are in agreement with those reported in other studies (Table 2). The fat content of the kernels of A. digitata ranged from 31.9 to 39.9 g/100 g DM, and in the kernels of S. birrea from 49.4 to 63.1 g/100 g DM. The protein contents in the kernels of A. digitata and S. birrea were high, 29.2 to 42.7 g/100 g DM. Similar, high protein contents have been found in other studies (Table 2). This means that the kernels are good sources of fat and protein.

33 Table 8. Nutritional composition of the pulp and kernels of the wild fruits from Mozambique determined in this study Fruit part Location Dry Fat Protein Dietary fibre Ash year matter Insoluble Soluble (g/100 g) (g/100 g DM) Pulp A. digitata Tete, 08 89.8±0.1 0.5±0.1 2.4±0.0 14.2±0.4 60.3±1.9 5.5±0.1

Tete, 09 89.1±0.0 0.7±0.6 2.2±0.1 14.7±0.0 65.6±0.6 7.4±0.0

Vilan, 09 86.5±0.1 0.5±0.1 2.1±0.0 16.1±0.8 57.3±0.3 7.4±0.0

L. kirkii Marr, 08 27.7±0.0 0.9±0.1 2.1±0.2 3.5±0.6 4.6±0.1 2.9±0.0 Marr, 09 23.9±0.0 0.4±0.2 n.a. 4.9±0.8 4.3±0.3 3.5±0.0 Man, 09 20.1±0.3 1.3±0.1 1.7±0.0 4.9±0.4 5.8±1.2 3.0±0.0

S. kraussii Marr, 08 16.4±0.3 0.8±0.5 3.7±0.0 3.3±0.1 6.0±0.4 4.8±0.2 Marr, 09 17.3±0.1 1.5±0.6 2.3±0.0 2.6±1.2 7.6±1.0 3.4±0.0 Man, 09 16.5±0.1 0.7±0.2 2.1±0.0 5.5±0.3 7.1±0.4 3.6±0.0

S. birrea Man, 09 16.8±0.1 0.9±0.3 1.4±0.1 7.7±1.7 10.5±0.9 3.0±0.1

V. infausta Marr, 08 37.4±0.5 0.5±0.8 2.9±0.2 45.6±1.4 24.3±1.4 7.8±0.6 Marr, 09 30.0±0.1 0.7±0.1 2.2±0.3 45.8±0.9 26.3±0.9 5.3±0.0 Man, 08 37.3±0.9 0.7±0.2 4.7±0.4 41.0±0.8 10.6±0.8 5.7±0.4

Man, 09 34.5±0.7 0.2±0.0 3.3±0.8 30.9±1.9 23.1±1.9 3.2±0.0 Kernels

A. digitata Tete, 08 93.6±0.0 35.0±0.2 36.7±0.9 17.4±3.4 14.0±0.6 7.7±0.0

Tete, 09 91.7±0.1 39.9±6.2 38.6±0.8 20.9±1.3 17.0±3.4 7.2±0.0

Vilan, 09 90.6±0.0 39.0±7.0 42.7±0.7 14.7±0.0 42.6±1.8 8.5±0.0

Tete ,13 92.8±0.0 31.7±0.0 35.0±0.0 n.a. n.a. n.a.

S. birrea Man, 08 93.6±0.0 58.3±1.3 30.1±0.2 17.6±2.9 10.5±2.7 3.8±0.0

Man, 09 95.0±0.0 63.1±0.1 35.0±0.1 18.5±1.5 8.4±2.4 3.5±0.0

Man, 13 95.8±0.1 49.4±3.7 29.2±0.0 n.a. n.a. n.a. n.a. = Not analysed; Vilan= Vilanculos; Marr=Marracuene, Man= Manhiça; 08 = 2008; 09 = 2009; 13 = 2013.

34 Table 8 also presents the results of the dietary fibre analysis. Large amounts of dietary fibre were found in the pulps of A. digitata and V. infausta. These contents are high compared with apples and guava in Table 1. The highest amount of soluble dietary fibre, around 60 g/100 g DM, was found in A. digitata pulp and the highest amount of insoluble dietary fibre, 31 to 46 g/100 g DM, in V. infausta pulp. The kernels of A. digitata and S. birrea contained large amounts of dietary fibre. The amounts found were higher than in previous studies (Table 2). However, the variations may be partly due to the use of different analytical methods in the various studies. The analysis of the dietary fibre composition showed that the contents of monosaccharides and uronic acids in the soluble and insoluble dietary fractions of the kernels differed between the kernels (Paper III). The main constituent of the insoluble fraction of A. digitata kernels was glucose, followed by arabinose and uronic acids, while in the soluble fraction, arabinose was the dominating component, followed by uronic acids. In kernels of S. birrea, uronic acids constituted more than 90% of both dietary fibre fractions. The ash content (Table 8), providing an indication of the mineral content, was around 3 to 9 g/100 g DM. The results generally agreed with those reported in previous studies (Table 2). The ash content of the fruits in Table 1 is below 0.7 g/100 g on wet weight. The pH, titratable acidity and soluble solids are important parameters for the food industry in evaluating the quality and maturity of fruits (Lammertyn et al., 2000). The pH of the pulps showed an acidic character (around pH 3), apart from S kraussii, the pH of which was slightly above 6. The acidic character is in accordance with previous data on pulps from A. digitata, S. birrea and V. infausta (Amarteifo and Mosase, 2006; Mothapo et al., 2014). The pH of fruits generally varies between 2.5 and 4.5 depending on their content of organic acids (Calvacanti et al., 2006). Titratable acidity is related to the acidic or sour taste, and is expressed as the amount of organic acids. The titratable acidity of the pulp, estimated as the percentage of citric acid, ranged from 0.6 to 1.7%. Corresponding values in the literature are 7.8% for A. digitata, 0.9% for S. birrea and 1.7% for V. infausta (Amarteifo and Mosase, 2006). The soluble solids content, measured as °Bx (degree Brix), is often used as an approximation of the concentration of sugars in the fruits (Silva et al., 2009). However, other soluble components, for example soluble dietary fibre, will also contribute to the degree Brix. The soluble solids content in the pulps ranged from 14 to 35°Bx; the results for fruits collected in Manhica in 2009 are presented in Table 9. A lower value (3.9°Bx) has been reported in the literature for V. infausta (Mothapo et al., 2014).

35 The contents of different mono- and disaccharides in the pulps were determined and the results, expressed as g/100 g wet weight, are presented in Table 9.

Table 9. Mono- and disaccharides in the pulps from wild fruits collected in Manhiça in 2009 Sample Glucose Fructose Sucrose Lactose Maltose Soluble solids

(g/100 g wet weight) (°Brix) A. digitata 3.0 3.0 4.3 <0.04 <0.04 n.a L. kirkii 7.5 5.7 1.2 <0.04 <0.04 26.3 S. kraussii 3.7 3.9 <0.1 <0.04 <0.04 15.0 S. birrea 0.5 0.4 1.4 <0.04 <0.04 17.4 V. infausta 1.4 1.4 2.7 <0.04 <0.04 30.1 n.a. = not analysed.

The highest amounts of glucose and fructose were found in L. kirkii, 13.2 g/100 g in total, and the lowest amount in S. birrea, below 1 g/100 g. The highest amount of sucrose was found in A. digitata, 4.3 g/100 g, and the lowest amount in S. kraussii, below 0.1 g/100 g. Lactose and maltose were detected in all samples, but at much lower concentrations, <0.04 g/100 g. No data could be found in the literature on the sugar contents of the studied fruits. When comparing the contents of individual sugars or the sum of mono- and disaccharides, no correlation was observed between the contents of sugars and the content of soluble solids (°Brix). The high content of soluble solids in V. infausta (30.1 °Brix) in spite of its rather low sugar content is explained by the high content of soluble dietary fiber (Table 9).

36 36 The same pulp samples were analysed with regard to organic acids. Organic acids are not routinely determined at the Department of Food Technology, Engineering and Nutrition. Therefore, a new method was developed based on information from the manufacturer of the ion chromatograph (See Paper II). The elution solvent and the column temperature were varied to find conditions that gave reliable results. The separation of citric, malic, tartaric and succinic acids was investigated, and the optimal conditions for their separation were found to be an 85:15 (v/v) mixture of 0.5 mM sulphuric acid and acetone, and a column temperature of 30°C The calibration curve was linear in the range studied. A chromatogram from the analysis of organic acids in S. kraussii is shown in Figure 8. The peaks corresponding to the different organic acids and two unknown compounds are indicated by numbers.

Figure 8. Chromatogram from the analysis of organic acids in S. kraussii. The peaks correspond to: citric acid (1), tartaric acid (2), malic acid (3) and succinic acid (4). (Peaks 5 and 6 were not identified.)

37 37 The results of the analysis of the organic acids, expressed in g/kg wet weight, are given in Table 10. Citric, malic and succinic acids were found at various amounts in all pulps, and traces of tartaric acid in one sample.

Table 10. Organic acid content in pulps from wild fruits collected in Manhiça in 2009

Sample Dry matter Citric acid Malic acid Succinic acid Tartaric acid

(g/100 g) (g/kg wet weight)

A. digitata 88.0 25.7±2.1 1.6±0.2 0.1± 0.0 n.d.

L. kirkii 18.0 21.5±2.1 0.4±0.1 0.03±0.0 n.d.

S. kraussii 9.0 0.9±0.3 1.1±0.1 0.1±0.0 n.d.

S. birrea 8.0 8.5±1.3 1.2±0.2 0.1±0.0 Trace

V. infausta 31.0 6.2±0.5 2.1±0.2 0.1±0.0 n.d. n.d. = Not detectable, Trace = < the determination limit (0.01 g/kg).

The maximum content of malic acid was 2.1 g/kg, and succinic acid was below 0.1 g/kg. High amounts of citric acid were found in A. digitata and L. kirkii; above 20 g/kg. When tasting the fruits, A. digitata and L. kirkii were the most acid ones. The results for A. digitata are in accordance with results in a previous study. Citric acid in A. digitata has been reported to be present at amounts around 23 to 32 g/kg, and malic acid 1.1 to 1.4 g/kg (PhytoTrade Africa, 2008). For comparison, the citric acid concentrations in some traditional fruits are: 2.2 g/ kg in pineapple, 4.5 g/ kg in oranges, 13.1 g/ kg in grapes and 41.2 g/ kg in limes (Falade et al., 2003). The presence of citric and malic acid in fruits may promote iron absorption (Gillooly et al., 1983).

38 38 The mineral contents of the fruits and kernels, expressed as mg/100 g DM, are given in Table 11. Although it is important, iodine was not included in this study as it is not commonly found in fruits (Hurrell, 1997).

Table 11. Mineral contents of the pulps, whole seeds and kernels of the studied wild fruits Fruit, part Dry Location, Ca K Fe Mg Na P S Se Zn matter year (g/100 g) (mg/100 g DM) Pulp A. digitata Tete, 08 90.1 326 2308 2.0 162 5.0 40 51 n.d. n.a. Tete, 09 86.4 308 2392 2.0 129 2.0 40 41 n.a. 0.5

Vilan, 09 86.5 366 2360 1.0 102 5.0 30 43 n.a. 0.6

L. kirkii Man, 09 20.4 28 1840 4.0 51 21 55 20 n.a. 1.4

S. kraussii Man, 09 8.9 127 2056 9.0 207 35 153 191 n.a. 0.6 S. birrea Man, 09 15.9 201 2753 3.0 138 30 178 90 n.a. 0.7 V. infausta Man, 09 34.4 90 1249 3.0 65 18 92 45 2.0 n.a.

Seeds A. digitata

Tete, 08 93.6 220 1238 29 360 2.0 523 131 n.d. n.a. Kernels A. digitata Tete, 09 92.2 293 1416 6.0 626 1.0 1229 263 n.a. 5.7

Vilan, 09 90.6 347 1451 4.0 706 2.0 1518 269 n.a. 5.2 S. birrea Man, 08 92.7 60 622 4.0 436 7.0 871 284 n.d. n.a. Man, 09 95.0 81 531 4.0 396 6.0 719 269 n.a. 4.5 n.d. = Not detectable; n.a. = Not analysed. Vilan=Vilanculos; Man=Manhiça; 08=2008; 09=2009.

40 39 The results can be compared with the literature data given in Table 3. The discussion below focuses on some of the minerals that are often deficient in local diets, i.e. calcium, iron and zinc. The highest amount of calcium was found in A. digitata pulp and kernels, about 300 mg/100 g DM or more. This value generally agrees with literature data (Table 3), but is lower than that reported in a previous study (Nour et al., 1980). The amount of iron in the pulps ranged from 1 to 9 mg/100 g DM; the highest amount being observed in S. kraussii. These amounts are comparable to those found in other studies (Table 3), however, higher amounts of iron have been reported in V .infausta pulp (24.4 and 28.3 mg/100 g DM), see Table 3. The highest iron content, 29 g/100 g DM, was found in the seeds of A. digitata, and was 5 to 7 times higher than in the kernels. Similar findings of higher amounts in the whole seed than in the kernels have been reported in the literature (Table 3). The zinc content in the pulps was around 1 mg/100 g DM or below. The zinc content in the kernels of A. digitata and S. birrea was around 5 mg/100 g DM. This is in agreement with results from a previous study of S. birrea kernels (see Table 3). Phytic acid may decrease the absorption of minerals from the diet. According to the literature, A. digitata and S. birrea kernels have been shown to contain phytic acid, for a literature review, see Table 4. Thus, preliminary experiments were performed to determine the content in the kernels. The results showed that kernels of A. digitata contained about 5.5% phytic acid and S. birrea kernels 3.4%. Comparing the results with the literature data given in Table 4, it can be seen that the A. digitata kernels had a lower content than in some previous studies (Mitchikpe et al., 2008; Adubiaro et al., 2011), and higher than in others (Osman, 2004; Ezeagu, 2005; Saulawa et al., 2014). The differences between the results may be explained to some extent by the method of analysis, which is sometimes not clearly described. The results regarding the amounts of phytic acid in S. birrea kernels are almost six times higher than that reported in a previous study (Muhammad et al., 2011). It is interesting to note that the phytic acid content in peanuts, which are commonly consumed in Mozambique, has been reported to be around 2 g/100 g (Lazarte et al., 2015). Results from additional experiments showed that the presence of phytic acid in the kernels can be reduced by various processing techniques (Paper IV). For example, treatment with phytase reduced the phytic acid content by 20-30% after only 15 minutes, and after 4 hours’ enzymatic incubation, the decrease was significant. Interestingly, almost 50% of the estimated original content of minerals was found in the supernatant after a few minutes’ enzyme incubation. This suggests that the water

40 used for boiling should be included in the dishes if the kernels are to be used as a dietary supplement.

41 The preliminary studies showed that the kernels had high protein content, and it was thus deemed interesting to determine the amino acid content. The results of the amino acid analysis are presented in Table 12.

Table 12. Amino acid composition in kernels of A. digitata and S. birrea Amino acid A. digitata S. birrea Peanuts1 (g/100g of protein)

Cysteine* 0.5 ± 0.0 0.7 ± 0.0 1.3 Histidine* 2.2 ± 0.1 2.5 ± 0.2 2.5 Isoleucine* 3.4 ± 0.2 3.9 ± 0.3 3.5 Leucine* 6.0 ± 0.4 4.3 ± 1.3 6.5 Lysine* 3.4 ± 1.0 2.9 ± 0.2 3.6

Methionine* 0.8 ± 0.1 0.7 ± 0.1 1.2 Phenylalanine* 4.5 ± 0.3 4.6 ± 0.3 5.2 Threonine* 2.8 ± 0.1 2.8 ± 0.2 3.4 Valine* 4.5 ± 0.3 5.0 ± 0.4 4.2 Alanine 3.8 ± 0.3 3.7 ± 0.3 3.9 Arginine 11.4 ± 0.6 13.9 ± 1.1 12.0 Aspartic acid 8.1 ± 0.4 8.4 ± 0.5 12.2 Glycine 4.4 ± 0.3 4.9 ± 0.3 6.0 Glutamic acid 23.7 ± 1.3 25.7 ± 1.8 20.8 Proline 3.2 ± 0.2 3.5 ± 0.2 4.4 Serine 4.6 ± 0.3 2.9 ± 0.3 4.9 Tyrosine 2.5 ± 0.1 4.7 ± 0.2 4.1 Total 89.8 95.9 99.7 * Essential amino acid 1Data recalculated from USDA, Nutrient Database for Standard Reference, release 27, accessed at 22/11/ 2014.

4442 All common amino acids, except tryptophan, were detected in both kernels. Under acidic hydrolysis conditions, tryptophan is totally destroyed, while asparagine and glutamine are hydrolysed quantitatively to aspartic acid and glutamic acid, respectively. Cysteine is readily oxidized to cysteic acid, and methionine to methionine sulphoxide, which is partly lost during hydrolysis. The amino acid profiles given in Table 12 generally agree with reports on amino acid contents in whole seeds and kernels (Table 5). However, the values for the S-amino acids are lower. This may be explained by partial breakdown of these amino acids during hydrolysis. Regarding the other essential amino acids, the results for lysine and valine in A. digitata are lower than in one previous study (Osman, 2004), and the findings regarding leucine, lysine and threonine in S. birrea are also lower (Muhammad et al., 2011). It is interesting to note that the amino acid composition in the two kernels studied is comparable to that in peanuts (WHO, 2007), which commonly form part of the diet in the locations from which the fruits were obtained.

4543 The results from the first study showed that the kernels had a high fat content, and the fatty acid composition was therefore determined. The results are presented in Table 13, together with the fatty acid compositions of olive oil and peanuts for comparison.

Table 13. Fatty acid composition of kernels of A. digitata and S. birrea Fatty acid A. digitata S. birrea Olive oil1 Peanuts1 No. of carbon atoms Systematic and name (g/100 g of fat) double bonds C14:0 Myristic 0.2 ± 0.0 n.d. - 0.4 C16:0 Palmitic 25.7 ± 0.2 12.1 ± 0.4 11.1 10.4 C16:1 Palmitoleic 0.2 ± 0.0 0.2 ± 0.0 1.2 0.0 C17:0 Margaric 0.2 ± 0.0 0.1 ± 0.0 - - C17:1 Margaroleic 0.2 ± 0.0 n.d. - - C18:0 Stearic 4.6 ± 0.1 7.3 ± 0.1 2.7 2.2 C18:1 Oleic 34.9 ± 0.7 72.4 ± 0.6 70.7 48.3 C18:2 Linoleic* 29.9 ± 0.9 6.8 ± 0.4 8.1 31.7 C18:3 Linolenic* 2.1 ± 0.0 n.d. 0.7 0.0 C20:0 Arachidic 0.9 ± 0.1 0.6 ± 0.0 - 0.0 C20:1 Eicosenoic 0.2 ± 0.0 0.3 ± 0.0 - - C22:0 Behenic 0.3 ± 0.0 0.1 ± 0.0 - - C22:1 Erucic 0.1 ± 0.0 n.d. - - C24:0 Lignoceric 0.2 ± 0.0 0.2 ± 0.0 - - 1Data recalculated from The Swedish National Food Administration’s food database, version 28/02/2014. n.d. = Not detectable; *Essential fatty acid.

The unsaturated fatty acids constituted about 68 and 80% of the total fat content in A. digitata and S. birrea kernels, respectively. The most abundant fatty acids found in A. digitata kernels were oleic acid (29.9%) and linoleic acid (34.9%); the latter being an essential fatty acid. Interestingly, A. digitata kernels also contained the essential fatty acid linolenic acid. In S. birrea kernels, oleic acid constituted 72.4% of the fatty acids. The results for A. digitata are in agreement with some previous studies but

4644 higher than in an older study (See Table 6). The results for S. birrea agree with most data in the literature (Table 6). The oils from both kernels have been reported to be very stable (Mariod et al., 2004; Zimba et al., 2005; Kleiman et al., 2008). It is interesting that the fatty acid composition in S. birrea is almost the same as in olive oil, with regard to the most abundant fatty acids; palmitic, oleic and the essential linoleic acid or A. digitata, the fatty acid composition is similar to that in peanuts, both with high contents of oleic and linoleic acid, but A. digitata also contains linolenic acid.

4745 46 5.3 Dietary intake estimations

An adequate diet requires that the food eaten can provide all the essential nutrients in sufficient amounts to support growth and maintain health (Whitney and Rolfes, 1996). Some authorities, e.g. the Food and Nutrition Board of the US Institute of Medicine, have established adequate intake (AI) levels for a number of different nutrients. Some results from the chemical analysis of the pulps and kernels were selected for the estimation of the intake in some vulnerable age groups. A portion size of 100 g fresh fruit or 40 g kernels was assumed in these estimates. The mineral content of the fruit pulps seemed high, but as the results are based on dry weight basis, the contribution of minerals from 100 g fresh fruit was estimated to be below 10% of the AI values. However, the A. digitata pulp had a high DM content, and consumption of 100 g would give around 29% of the AI of calcium for pregnant women and 30% for children and also 23% of the AI of iron for 4 to 13 years old (data not shown). The kernels, with a high DM content, may contribute considerably to the AI. Assuming a consumption of 40 g kernels, estimates were made for some selected age groups and are presented in Table 14.

Table 14. Estimated contribution of minerals through the consumption of 40 g kernels A. digitata S. birrea Mineral Life stage group AI1 Estimated contribution (g/day) (% of AI) Calcium Children 4 to 8 years 0.8 16 5 Older children 9 to 13 years 1.3 10 5 Pregnant women 1.0 13 5 (mg/day) Iron Children 4 to 8 years 10 20 16 Older children 9 to 13 years 8 25 20 Pregnant women 27 7 6 Zinc Children 4 to 8 years 5 44 36 Older children 9 to 13 years 8 27 23 Pregnant women 11 20 16

1Adequate intake according to the Food and Nutrition Board, 2002.

48 47 Calcium is important for the growth of bone and teeth (Insel et al., 2011). It can clearly be seen that consumption of 40 g A. digitata kernels would provide 10 to 16% of the AI of calcium for the selected groups, while the contribution from S. birrea would be below 5%. Iron plays a role in the transport of oxygen in the body and has many other functions (Insel et al., 2011). Consumption of 40 g A. digitata kernels can provide 20 to 25% of the AI for children and S. birrea kernels 16 to 20%. For pregnant women, the contribution is considerably lower. Interestingly, ground whole seeds (shell + kernel) of A. digitata showed an extremely high iron content, and the contribution from 40 g could provide more than 100% of the AI for children and 43% for pregnant women (data not shown). For zinc, the consumption of 40 g A. digitata kernels would provide 20 to 44% of the AI for the selected groups, while S. birrea kernels would provide slightly less, 16 to 36%. Zinc has a number of functions, for example, the body needs zinc to grow and develop properly during pregnancy, infancy and childhood (Brown et al. 1999; Insel et al. 2011). Similar calculations were performed for protein. The protein content in the fruit pulp was low, and thus the contribution from pulp will be small. The results for the kernels are presented in Table 15.

Table 15. Estimated contribution of protein through the consumption of 40 g kernels A. digitata S. birrea

Life stage group AI1 Estimated contribution (g/day) (% of AI) Children 4 to 8 years 19 81 66

Older children 9 to 13 years 34 45 37

Pregnant women 71 22 18

1Adequate intake according to the Food and Nutrition Board, 2002.

For children aged 4 to 8 years, 81 and 66% of the AI is covered by the consumption of 40 g A. digitata kernels and S. birrea kernels, respectively. For older children and pregnant women, the contribution is lower, but it may be possible for these groups to increase their intake of kernels, especially if they are eaten as a snack. Furthermore, these two groups generally eat larger portions of dishes containing protein.

48 49 Ideally, the protein in the diet should provide the body’s requirement of all the essential amino acids, in the appropriate relative proportions. The content of essential amino acids in relation to need is a measure used to determine the protein quality. The contents of essential amino acids in the kernels were compared with the amino acid requirements stated by the World Health Organization (WHO) for children aged 3-10 years; the results are presented in Table 16. It should be noted that, according to the WHO, the sum of phenylalanine and tyrosine is important, although tyrosine is a non-essential amino acid. Histidine is regarded as being essential for children, but not for adults.

Table 16. Essential amino acids in kernels of A. digitata and S. birrea compared with the WHO protein requirement for children aged 3-10 years old Amino acids A. digitata S. birrea WHO, 2007

(g/100 g protein) Histidine 2.2 2.5 1.6 Isoleucine 3.4 3.9 3.0 Leucine 6.0 4.3 6.1 Lysine 3.4 2.9 4.8 Methionine + cysteine 1.3 1.4 2.3 Phenylalanine + tyrosine 7.0 9.3 4.1 Threonine 2.8 2.8 2.5 Valine 4.5 4.9 4.0

The total content and relative amounts of the different essential amino acids are generally similar to or above that recommended by the WHO. However, the amounts of methionine + cysteine, and lysine in both kernels, and also leucine in S. birrea kernels, were lower than required. The amount of the sulphur-amino acids were lower because they are partly destroyed during acid hydrolyses in the sample prepartion. The total amino acid composition indicates that A. digitata and S. birrea kernels can provide good, cheap sources of protein, especially if combined with foods with higher contents of lysine.

50 49 Fat is another important compound in the diet. The Food and Nutrition Board has established AI values in g/day for omega-3 and omega-6 fatty acids. The European Food Safety Authority (EFSA) and the WHO, on the other hand, recommend an acceptable range of omega-3 and omega-6 fatty acid intake related to the energy intake. Based on the results of the analysis of the fatty acid composition in the two kernels, the contributions of omega-3 and omega-6 fatty acids from the consumption of 40 g kernels were estimated. The results are presented in Table 17.

Table 17. Intake of omega-6 and omega-3 fatty acids through the consumption of 40 g of kernels A. digitata S. birrea Fatty acids Life stage group AI1 Estimated contribution (g/day) (% of AI) Omega-6 fatty acids Children 4 to 8 years 10 120 27 (Linoleic acid) Males 9 to 13 years 12 100 23 Females 9 to 13 years 10 120 27 Pregnant women 13 92 21 Omega-3 fatty acids Children 4 to 8 years (α-Linolenic acid) 0.9 93 - Males 9 to 13 years 1.2 70 - Females 9 to 13 years 1.0 84 - Pregnant women 1.4 60 - 1Adequate intake according to the Food and Nutrition Board, 2002.

It can be seen that 40 g A. digitata kernels can cover the daily intake of omega-6 fatty acids for those aged 4 to 13 years, and about 90% of the requirement for pregnant women. The same quantity of A. digitata kernels can provide 60 to 93% of the daily requirement of omega-3 fatty acids for the same groups. Omega-3 fatty acids are not found in peanuts, which often form a considerable part of the diet in rural areas. Thus the A. digitata kernels may be an important and nutritious ingredient in different dishes. The contribution of omega-6 fatty acids from 40 g S. birrea kernels is 21 to 27% of the AI for selected groups, and if S. birrea kernels are the only source of omega-6 fatty acids, an intake of around 160 g kernels is needed. Omega-3 fatty acids were not detected in S. birrea kernels, however, these kernels may not be the only source of essential fatty acids in the diet.

50 51 The intake of dietary fibre has been related to various health effects, appear to be at significantly lower risk for developing coronary heart disease, diabetes, obesity, blood pressure, cholesterol levels and certain gastrointestinal diseases (Buttriss and Stokes, 2008). Assuming consumption of 100 g pulp of V. infuasta, 40 g of A. digitata pulp, 40 g of A. digitata and S. birrea kernels, estimates were made for selected life stage groups and are presented in Table 18.

Table 18. Estimated contribution of dietary fibre through the consumption 40 g kernels Life stage group AI1 Estimated contribution (% of AI) (g/day) V. infausta A. digitata A. digitata S. birrea 100 g Pulp 40 g Pulp 40g Kernels 40 g Kernels

Females 19 to50 years 25 86 108 61 42

Males 14 to 50 years 35 61 71 44 30

1Adequate intake according to the Food and Nutrition Board, 2002.

Consumption of 100 g pulp of V. infausta could provide around 61 to 86% of the AI of dietary fibre for adults and consumption of 40 g A. digitata pulp could provide 71 to more than 100%. The contribution from 40 g kernels is lower but is estimated to cover around 30 to 61% of the AI.

52 51 52 6. Conclusions and Future Outlook

The results of this research present a contribution to bridge the gap between traditional understanding and scientific knowledge of the wild fruits studied. The fruits are commonly consumed by those living in most districts of Mozambique, especially by children, and form part of their normal diet. Also, the kernels of A. digitata and S. birrea are usually consumed in rural areas. Wild fruit trees have significant cultural and socio-economic value in the countryside of Mozambique. Most of the wild fruits are seasonal, and are available mainly in the wet season. They generally have a short shelf-life and are eaten fresh or after minimal processing. The most common method of preservation is sun-drying. Although there is little documented information on the composition and nutritional benefits of these fruits, rural residents believe in, and use, traditional knowledge passed down from their forefathers. The nutritional components and other characteristics have been determined for the five fruits studied in this work. No data could be found in the literature for L. kirkii and S. kraussii, and data on S. birrea and V. infausta were scarce. Small, non- significant variations were observed between fruits gathered from different locations and in different years, which may be due to differences in climate, soil and weather conditions. The pH, titratable acidity and the contents of soluble solids, sugars and dry matter in the fruit pulps were determined, as these are interesting for the fruit processing industry. They are of importance for taste, aroma and the prevention of spoilage by microorganisms. The A. digitata pulp had remarkably high dry matter content, around 85 to 90%, compared with pulps from the other fruits. Therefore, this fruit has a long shelf-life. The total amount of glucose and fructose found in L. kirkii was about 13 g/100 g fresh weight. The highest amount of sucrose was found in A. digitata, about 4 g/100 g, while the content of these sugars was below 2 g/100 g in S. kraussii. No substantial amounts of lactose or maltose were detected in the samples. The protein content was low in all the fruit pulps, as is the case in fruits in general. However, the protein content was high in the kernels of A. digitata and S. birrea, around 30 to 40% on dry matter basis. The amino acid composition in the kernels was generally good according to comparison with the recommendations for amino acid requirements given by the WHO. This means that the kernels can provide good, cheap sources of protein.

5553 The fat content was below 2% in the fruit pulps, while the fat content in A. digitata kernels was almost 40%, and in and S. birrea kernels, about 60%. The unsaturated fatty acids constituted about 70 and 80% of the total fat content in A. digitata and S. birrea kernels, respectively. Interestingly, the A. digitata kernels contained the essential fatty acids linoleic and linolenic acid, while the S. birrea kernels contained only linolenic acid. All the studied fruit pulps and kernels contained dietary fibre. The pulp of A. digitata had the highest amount of soluble dietary fibre, about 60%, while the V. infausta pulp had the highest amount of insoluble dietary fibre, about 40%. The content of dietary fibre was lower in the kernels. Analysis of the mineral content showed that both pulp and kernels from A. digitata contained an appreciable amount of calcium: more than 300 mg/100 g on dry matter basis. The whole seeds (shell + kernel) of A. digitata contained as much as 29 mg iron per 100 g. Interestingly, the S. kraussii pulp had a high iron content, 9 mg/100 g. The kernels of A. digitata and S. kraussii had high contents of magnesium, about 600 and 400 mg/100 g, respectively. Comparison with recommended adequate intake shows that the kernels may help to meet the mineral requirements of children and pregnant women in rural areas. A new analytical method was developed for the determination of organic acids in the fruit pulps. Citric and malic acids were found at various amounts in all pulps together with small amounts of succinic acid. Tartaric acid was not detected in the samples. Large amounts of citric acid were found in A. digitata and L. kirkii, above 20 g/kg fresh weight. When tasting the fruits, A. digitata and L. kirkii were the most acid ones. S. kraussii had the lowest total content of these organic acids, below 2.2 g/kg.

Phytic acid, which may decrease the absorption of minerals from the diet, was detected in kernels of A. digitata and S. birrea at concentrations of about 5.5% and 3.4%, respectively. Incubation with phytase reduced the phytic acid content by 20 to 30% after only 15 minutes. Interestingly, almost 50% of the estimated original content of minerals was found in the supernatant after a few minutes’ incubation. This suggests that the water used for boiling should be included in the dishes if the kernels are to be used as a dietary supplement. The findings of this study show that A. digitata, S. birrea and V. infausta pulps can provide large amounts of nutrients, such as minerals and dietary fibre. The kernels of A. digitata and S. birrea, may help to supply fat, protein and various minerals. A value addition to the wild fruits could be achieved by the development of small scale methods for processing, either individually or in combination with fruits with

5654 different nutrient content or with cereals abundant in the areas. Thus, there is still a great need to continue the chemical analysis of other type of wild fruits that may make a significant contribution to a nutritious diet.

The new data obtained from this study will be vital in efforts to promote the greater use of wild fruits and their kernels, for the estimation of dietary intakes, and the education of local communities with regard to the nutritional benefits of free sources of food in their environment. The findings can also contribute to the domestication of wild fruits trees, and the enhancement of cultural heritage and forests. In addition, the nutritional analysis of the investigated wild fruits can form the basis for the development of further processing methods to extend shelf-life and for the production of other food products.

5755 Acknowledgements

Many people have contributed to this work, all of whom have been important to me and deserve special thanks.

I would first like to express my gratitude to my main supervisor Kerstin Skog for taking me on as her PhD student. I appreciate your inspiration, help and guidance, the scientific discussions and I learned a lot from you. I also really appreciate the time I spent with your all family, the trips out of Lund to Karlshamn and the islands, the Mörrum River, dinners…and so one… are pleasant and unforgettable moments !! Thank you so much. I would like to express my special thanks to my co-supervisors Ingegerd Sjöholm and Amália Uamusse, thanks for the invaluable collaboration, scientific discussions, encouragement and support, which enabled me to pursue and complete this research. Ingegerd thanks a lot for the pleasant boat trip to the small islands and many social events that contributed to the joyful atmosphere. Estera Dey, Björn Bergenståhl and Yvonne Granfeldt, thank you for the scientific discussions, encouragement and support, and for contributing to the positive atmosphere in Lund. Louis Pelembe, at the Eduardo Mondlane University in Mozambique, thank you for introducing me to research in wild fruits. Everyone at the Department of Food Technology, Engineering and Nutrition, especially my colleagues Borges, Eulália, Irene, Lucas and Raice from Mozambique, Claudia, Deysi, Randi, Liyana and Kataneh for their contributions and all the good moments we had together. Bindu, Cristina, Diana, Tannaz, Thao, Yadong, Yoga and Yonna, for sharing knowledge, pleasant friendly atmosphere, especially at lunch times.

Dan Johnson, Lina Haskå, Lisbeth Person, Ulf Nilsson, Christer Fahlgren, Serafina Vilanculos and Annette Almgren for excellent technical assistance. To my compatriots and colleagues at the Departments of Organic Chemistry and Water Resources at LTH, thank you for the friendly atmosphere. I also would like to thank Carlos Lucas and José da Cruz Francisco at the TecPro research programme in Mozambique, for encouragement and support during the research, the Financial Department on campus, and the Faculty of Engineering at Eduardo Mondlane University, and Orton Malipa, Benedito, Chissano and Guirrugo

56 for the efforts and flexibility in solving all the problems related with field trips and the travel involved during the period of my PhD studies.

Everyone at the Departments of Chemical Engineering and Biological Sciences at Eduardo Mondlane University in Mozambique, who played important roles at various stages of my research. Thank you so much.

Thanks, to the Centro Regional de Ciência e Tecnologia in Tete province, in particular Ana da Graça, Cardoso and Trindade, for help during sample collection and stays in Tete. To the residents of Bobole, Manhiça district and Tete province, for sharing traditional knowledge on the consumption and use of wild fruits, thank you. Thanks to Tomás Malo and his family.

Special thanks my family:

Agradeço a Deus, por ter cuidado de mim, proporcionando saúde e boa disposição durante este percuso! To my sisters Luciana, Milú, Epifânia, Mequilina and Palmira, my brothers-in-law Alberto and Clarêncio, all my nephews and in-laws: thank you for taking care of my family during my absence, especially my youngest daughter. To my husband Silva Magaia, my daughters Lutea and Shena, my son-in-law Atílio and my wonderful grandson Thanzi, for their patience and moral support throughout this time: I love you all from the bottom of my heart.

This work was supported by the Swedish International Development Cooperative Agency.

57 References

Abdul-Hamid, A., and Luan, Y.S. (2000). Functional properties of dietary fibre prepared from defatted rice bran. Food chemistry, 68(1), 15-19. Adubiaro, H.O., Olaofe, O., Akintayo, E.T., and Babalola, O.O. (2011). Chemical composition, calcium, zinc and phytate interrelationships in baobab (Adansonia digitata) seed flour. Advance Journal of Food Science and Technology, 3 (4), 228-232. Albano, G., and Nhamirre, G. (2007). Produtos florestais não madeireiros (PFNM) utilização e potencial para seu desenvolvimento em Moçambique. Ministério de Agricultura, República de Moçambique, report, written in Portuguese. Al-Farsi M, Alasalvar C, Morris A, Baron M, Shahidi, F. (2005). Compositional and sensory characteristics of three native sun-dried date (phoenix dactylifera L.) varieties grown in Oman. Journal of Agricultural and Food Chemistry, 53(19), 7586-7591. Amarteifio, J.O., and Mosase, M.O. (2006). The chemical composition of selected indigenous fruits of Botswana. Journal of Applied Sciences and Environmental Management, 10, 43-47. Anastasio, M., Pepe, O., Cirillo, T., Palomba, S., Blaiotta, G., and Villani, F. (2010). Selection and use of phytate-degrading LAB to improve cereal-based products by mineral solubilization during dough fermentation. Journal of Food Science, 75, M28-M35. Anderson, J.M., Smith, B.M., Gustafson N.J. (1994). Health benefits and practical aspects of high-fiber diets. American Journal of Clinical Nutrition, 59(suppl), 1242S-1247S. AOAC. (1980). Methods of Analysis of the Association of Official Analytical Chemists, 13rd edition, Association of Official Analytical Chemists, Washington, DC. AOAC. (1996). Methods of Analysis of the Association of Official Analytical Chemists, 16th edition, Association of Official Analytical Chemists, Washington, DC. AOAC. (2000). Methods of Analysis of the Association of Official Analytical Chemists, 17th edition, Association of Official Analytical Chemists, Washington, DC. Aphane, J., Chadha, M.L., and Oluoch, M.O. (2002). Increasing the consumption of micronutrient-rich foods through production and promotion of indigenous foods. FAO- AVRDC International Workshop (Proceedings), Arusha, Tanzania, 5-8 March 2002. Asp, N.-G., Johansson, C.-G., Hallmer, H., and Siljeström, M. (1983). Rapid enzymatic assay of insoluble and soluble dietary fiber. Journal of Agricultural and Food Chemistry, 31, 476-482. Boletin do Instituto de Investigação Agrária de Moçambique (IIAM). (2008). Ministério da Agricultura, Edição Trimestral, Janeiro-Março de 2008 N.º7, written in Portuguese.

Bonnet, B. (2006). Natural products: the new engine for African trade growth. Consultancy to further develop the trade component of the natural resources enterprise programme

6058 NATPRO. http://earthmind.net/biotrade/docs/sa/iucn-natpro-trade.pdf, accessed at 2015-02- 20. Borochov-Neori, H., Judeinstein, S., Greenberg, A., Fuhrman, B., Attias, J., Volkova, N., Hayek, T., Aviram, M. (2008). Phenolic antioxidants and antiatherogenic effects of marula (Sclerocarrya birrea Subsp. caffra) fruit juice in healthy humans. Journal of Agricultural and Food Chemistry, 56 (21), 9884-9891. Brown, L., Rosner, B., Willett, W.W., Sacks, F.M. (1999). Cholesterol-lowering effects of dietary fiber: a meta-analysis1, 2. American Journal of Clinical Nutrition, 69, 30-42. Buchmann, C., Prehsler, S., Hartl, A., Vogl, C.R. (2010). The Importance of Baobab (Adansonia digitata L.) in rural West African subsistence - suggestion of a cautionary approach to international market export of baobab fruits. Ecology of Food and Nutrition, 49 (3), 145- 172. Buttriss, J.L., and Stokes, C.S. (2008). Dietary fibre and health: an overview. British Nutrition Foundation Nutrition Bulletin, 33, 186-200. Calvacanti, A.L., Oliveira, K.F., Paiva, P.S., Rebelo Dias, M.V., Costa, S.K.P., and Vieira, F. F. (2006). “Determination of total soluble solids content (Brix) and pH in milk drinks and industrialized fruit juices,” Pesquisa Brasileira em Odontopediatria e Clíınica Integrada, 6 (1), 57-64. Carlsson, N.G., Bergman, E.L., Skoglund, E., Hasselblad, K., Sandberg, A.S. (2001). Rapid analysis of inositol phosphates. Journal of Agricultural and Food Chemistry, 49 (4), 1695-1701. Coultate, T.P. (2009). Food: The Chemistry of its Components, 5th edition. Royal Society of Chemistry, Cambridge UK, 501 pages. Commission Decision. (2008). Authorising the placing on the market of baobab dried fruit pulp as a novel food ingredient under regulation (EC) No 258/97 of the European Parliament and of the Council. http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2008:183:0038:0039:EN:PDF, accessed at 2012-06-05. Coughlin, P.E. (2006). Agricultural intensification in Mozambique. Infrastructural, policy and institutional framework-When do problems signal opportunities? Maputo, Mozambique. http://www.sarpn.org/documents/ 0002233/Agric_intensification_Coughlin_Sept2006.pdf, accessed at 2012-09-19. Da Silva, M.C., Izidine, S., and Amude, A.B. (2004). A preliminary checklist of the vascular plants of Mozambique. Southern African Botanical Diversity Network Report Nº 30. SABONET, Pretoria.

6159 De Caluwé, E., De Smedt, S., Assogbadjo, A.E., Samson, R., Sinsin, B., and Van Damme, P. (2009). Ethnic differences in use value and use patterns of baobab (Adansonia digitata L.) in northern Benin. African Journal of Ecology 47, 433-440. De Carvalho, M.F. (1968). Plantas silvestres de Moçambique com interesse alimentar. Gazeta do Agricultor, Moçambique. Separatas, 49, C., written in Portuguese. Emmanuel, T.V., Njoka, J.T., Catherine, L.W., and Lyaruu, H.M. (2011). Nutritive and anti-nutritive qualities of mostly prefereed edible woody plants in selected drylands of Iring District, Tanzania. Pakistan Journal of Nutrition, 10 (8), 786-791. Eteshola, E., and Oraedu, A. (1996). Fatty acid compositions of tigernut tubers (Cyperus esculentus L.), baobab seeds (Adansonia digitata L.), and their mixture. Journal of the American Oil Chemists' Society, 73, 255-257. Ezeagu, I.E. (2005). Baobab (Adansonia digitata L.) seed protein utilization in young albino rats I: Biochemical ingredients and performance characteristics. Animal Research International, 2(1), 240-245. Ezeagu, I., Petzke, K., Lange, E., and Metges, C. (1998). Fat content and fatty acid composition of oils extracted from selected wild-gathered tropical plant seeds from Nigeria. Journal of the American Oil Chemists' Society, 75, 1031-1035. EFSA. (2010). Scientific opinion on dietary reference values for fats, including saturated fatty acids, polyunsaturated fatty acids, monounsaturated fatty acids, trans fatty acids, and cholesterol, 8, 1459, http://www.efsa.europa.eu/en/efsajournal/pub/1461.htm, accessed at 2015-04-28. Effiong, G.S., and Udo, I.F. (2010). Nutritive values of four indigenous wild fruits in Southeastern Nigeria. Electronic Journal of Environmental, Agricultural and Food Chemistry, 9, 1168-1176. Eromosele, I.C., Eromosele, O., and Kuzhkuzha, D.M. (1991) Evaluation of mineral elements and ascorbic acid contents in fruits of some wild plants. Plant Foods for Human Nutrition, 41, 151-154. Eromosele, I.C., and Eromosele, C.O. (1993). Studies on the chemical composition and physico-chemical properties of seeds of some wild plants. Plant Foods for Human Nutrition, 43, 251-258. Elleuch, M., Bedigian, D., Roiseux, O., Besbes, S., Blecker, C., and Attia, H. (2011). Dietary fibre and fibre-rich by-products of food processing: Characterisation technological functionality and commercial applications: A review. Food Chemistry, 124(2), 411-421. Falade, O.S., Olusoga, R.S., Oladipo, A., Tubosun, A., Adewusi, S.R.A. (2003). The level of organic acids in some Nigerian fruits and their effect on mineral availability in composite diets. Pakistan Journal of Nutrition, 2 (2), 82-88.

6260 FAO. (2010). Crop and Food Security Assessment Mission to Mozambique. Agriculture Organization of the United Nations. Rome. http://www.fao.org/docrep/012/ak350e/ak350e00.pdf, accessed at 2012-09-10. FAO. (2011). For Improved Nutrition and Food Security. Food and Agriculture Organization of the United Nations. Rome. http://www.fao.org/forestry/27976-02c09ef000fa99932eefa37c22f76a055.pdf, accessed at 2012-09-10. FAO. (2012). Integrating Food Security Information in National Statistical systems. Food and Agriculture Organization of the United Nations. Rome. http://www.fao.org/docrep/015/i2588e/i2588e.pdf Accessed 2012-09-10. FDA. (2009). Agency Response Letter GRAS Notice No. GRN 000273. http://www.fda.gov/Food/FoodIngredientsPackaging, accessed at 2012-11-07. FAO. (2014). Food and Nutrition in Numbers, Food and Agriculture Organization of the United Nations, Rome. http://www.fao.org/publications/card/en/c/9f31999d-be2d-4f20-a645-a849dd84a03e/, accessed at 2015-02-27

FAOSTAT. (2014). Country profiles: Mozambique. http://faostat.fao.org/site/666/default.aspx, 2015-02-27. Food and Nutrition Board. (2002). Dietary fats: Total fat and fatty acids. Dietary reference intakes for energy, carbohydrate, fiber, fat, fatty acids, cholesterol, protein, and amino acids. In.:http://www.nap.edu/books/0309085373/html/, accessed at 2015-03-15. Gillooly, M., Bothwell, T.H., Torrance, J.D., MacPhail, A.P., Derman, D.P., Bezwoda, W.R., Mills, W., Charlton, R.W., and Mayet, F. (1983). The effects of organic acids, phytates and polyphenols on the absorption of iron from vegetables. British Journal of Nutrition, 49, 331-342. Glew, R. H., Dorothy J. VanderJagt, Cassius Lockett, Louis E. Grivetti, Garrett C. Smith, Andrzej Pastuszyn, and Millson, a. M. (1997). Amino acid, fatty acid, and mineral composition of 24 indigenous plants of Burkina Faso1. Journal of Food Composition and Analysis 10 205-217. Glew, R. S., VanderJagt, D. J., Huang, Y. S., Chuang, L. T., Bosse, R., and Glew, R. H. (2004). Nutritional analysis of the edible pit of Sclerocarya birrea in the Republic of Niger (daniya, Hausa). Journal of Food Composition and Analysis 17, 99-111. Greiner, R., and Konietzny, U. (1998). Endogenous phytate-degrading enzymes are responsible for phytate reduction while preparing beans (Phaseolus vulgaris). Journal of food processing and preservation, 22, 321-331. Haskå, L., Nyman, M., and Andersson, R. (2008). Distribution and characterisation of fructan in wheat milling fractions. Journal of Cereal Science, 48(3), 768-774.

6361 Hamilton, C. C., Geil, P. B. and, Anderson, J. W. (1992). Management of obesity in diabetes mellitus. Diabetes Education, 18(5), 407-410. Hassan, L.G., Dangoggo, S.M., Hassan, S.W., Muhammad, S., and Umar, K.J. (2010). Nutritional and antinutritional composition of Sclerocarya birrea fruit juice. Nigerian Journal of Basic and Applied Sciences, 18(2). Hernández, Y., Lobo, M. G., and González, M. (2006). Determination of vitamin C in tropical fruits: A comparative evaluation of methods. Food Chemistry, 96, 654-664. Hillman, Z., Mizrahi, Y., Beit-Yannai, E. (2008). Evaluation of valuable nutrients in selected genotypes of marula (Sclerocarya birrea ssp. caffra). Scientia Horticulturae, 117 (4), 321-328. Hoe, V.B., and Siong, K.H. (1999). The nutritional value of indigenous fruits and vegetables in Sarawak. Asia Pacific Journal Clinic Nutrition, 8, 24-31. Hurrell, R.F. (1997). Bioavailability of iodine. European journal of clinical nutrition, 51(1), S9- S12. Ibnouf, F.O. (2012). The value of women's indigenous knowledge in food processing and preservation for achieving household food security in rural Sudan. Journal of Food Research, 1 (1), 238-253. Igboeli, L.C., Addy, E.O.H., and Salami, L.I. (1997). Effects of some processing techniques on the antinutrient contents of baobab seeds (Adansonia digitata). Bioresource Technology, 59, 29-31. Imelouane, B., Tahri, M., Elbastrioui, M., Aouinti, F., and Elbachiri, A. (2011). Mineral contents of some medicinal and aromatic plants growing in eastern morocco. Journal of Materials and Environmental Science, 2(2), 104-111 Insel, P., Ross, D., McMahon, K., and Bernstein, M. (2011). Nutrition, Sudbury Massachusetts, 4th edition. Jones and Bartlett Publishers, USA. Jama, B.A., Mohamed, A.M., Mulatya, J., and Njui, A.N. (2008). Comparing the ‘‘Big Five’’: A framework for the sustainable management of indigenous fruit trees in the drylands of East and Central Africa. Ecological Indicators, 8, 170-179. Kleiman, R., Ashley, D., and Brown, J. (2008). Comparison of two seed oils used in cosmetics, moringa and marula. Industrial Crops and Products, 28(3), 361-364. Kaboré, D., Sawadogo-Lingani, H., Diawara, B., Compaoré, C.S., Dicko, M.H., and Jakobsen, M. (2011). A review of baobab (Adansonia digitata) products: Effect of processing techniques, medicinal properties and uses. African Journal of Food Science, 5(16), 833-844. Kennedy, G., Nantel, G., Shetty, P. (2003). The scourge of" hidden hunger": global dimensions of micronutrient deficiencies. Food Nutrition and Agriculture. 32, 8-16.

6462 Lammertyn, J., Peirs, A., De Baerdemaeker, J., and Nicolaý, B. (2000). Light penetration properties of NIR radiation in fruit with respect to non-destructive quality assessment. Postharvest Biology and Technology, 18,121-132. Lazarte, C.E., Carlsson, N-G., Almgren, A., Sandberg, A-S., and Granfeldt, Y. (2015). Phytate, zinc, iron and calcium content of common Bolivian food, and implications for mineral bioavailability. Journal of Food Composition and Analysis, 39, 111-119. López, M.A.A., and Martos, F.C. (2004). Iron availability: An updated review The International Journal of Food Sciences and Nutrition, 55(8), 597-606. Leakey, R.R.B. (1999). Potential for novel food products from agroforestry trees: a review. Food Chemistry, 66, 1-14. Lockett, C.T., Calvert, C.C., and Grivetti, L.E. (2000). Energy and micronutrient composition of dietary and medicinal wild plants consumed during drought. Study of rural Fulani, Northeastern Nigeria. International Journal of Food Science and Nutrition, 51, 195- 208. Manfredini, S., Vertuani, S., Braccioli, E., and Buzzoni, V. ( 2002). Antioxidant capacity of Adansonia digitata fruit pulp and leaves. Acta Phytotherapeutica, V, 2, 1-6. Mangue, P.D., and Oreste, M.N. (1999). Data collection and analysis for sustainable forest management in ACP countries-linking national and international efforts. EC-FAO Partnership programme (1998-2000), Report, Pp1-34. Maundu, P.M., Ngugi, G.W., and and Kakule, C.H.S. (1999). Traditional food plants of Kenya, Nairobi-Kenya, http://www.fastonline.org/CD3WD_40/CD3WD/AGRIC/H1093E/EN/B567_7.HTM, accessed at 2012-02-27. Murray, S.S., Schoeninger, M.J., Bunn, H.T., Pickering, T.R., and Marlett, J.A. (2001). Nutritional composition of some wild plant foods and honey used by Hadza foragers of Tanzania. Journal of Food Composition and Analysis, 14, 3-13.

Muhammad, S., Hassan, L.G., Dangoggo, S.M., Hassan, S.W., Umar, K.J., and Aliyu, R.U. (2011). Nutritional and antinutritional composition of Sclerocarya birrea seed kernel Studia Universitatis “Vasile Goldiş”, Seria Ştiinţele Vieţii 21 (4), 693-699, www.studiauniversitatis.ro, accessed at 2015-02-24. Mariod, A., Matthäus, B., and Eichner, K. (2004). Fatty acid, tocopherol and sterol composition as well as oxidative stability of three unusual Sudanese oils. The Journal of Food Lipids, 11(3), 179-189. Mitchikpe, E.C.S., Dossa, R.A.M., Ategbo, E.A.D., Van Raaij, J.M.A., Hulshof, P.J.M., and Kok, F.J. (2008). The supply of bioavailable iron and zinc may be affected by phytate in Beninese children. Journal of Food Composition and Analysis 21, 17-25.

6563 Mothapo, M.J., Mafeo, T.P., and Mamphiswana, N.D. (2014). Physico-chemical properties and selected nutritional components of wild medlar (Vangueria infausta) fruit harvested at two harvesting times. World Journal of Dairy & Food Sciences, 9 (1), 79-85. Motlhanka, D. M. T., Motlhanka, P., and Selebatso, T. (2008). Edible indigenous wild fruits plants of Eastern Botswana. International Journal of Poultry Science, 7 (5), 457-460. MBB Consulting. (2006). Report on A Study into the Potential of Aromatic Plants for Essential Oils in Mozambique. MBB Consulting Services South (Pty) Ltd Stellenbosch, South Africa In association with African Business Access Cape Town, South Africa. Project No. J1002/2, Reg. No: 2005/009287/07. www.portaldogoverno.gov.mz/.../Relatorio%20plantas%20aromaticast.p, accessed at 2015- 03-23. National Research Council. (2008). Lost crops of Africa: Volume III: Fruits. The National Academies Press, Washington, http://www.nap.edu, accessed at 2014-02-19. National Food Agency. (2012). The National Food Administration’s food database. SLV, Sweden, http://www.slv.se/en-gb/, accessed at 2012-09-11. Nkafamiya, I.I., Osemeahon, S.A., Dahiru, D., and Umaru, H.A. (2007). Studies on the chemical composition and physicochemical properties of the seeds of baobab (Adasonia digitata). African Journal of Biotechnology, 6, 756-759. Nnam, N.M., and Obiakor, P.N. (2003). Effect of fermentation on the nutrient and antinutrient composition of baobab (Adansonia digitata) seeds and rice (Oryza sativa) grains. Ecology of Food and Nutrition, 42, 265-277. Nour, A.A., Magboul, B.I., and Kheiri, N.H. (1980). Chemical composition of baobab fruit (Adansonia digitata L.). Tropical Science, 22, 383-388. Obizoba, I.C., and Amaechi, N.A. (1993). The effect of processing method on the chemical composition of baobab (Adansonia digitata. L) pulp and seed. Ecology of Food and Nutrition, 29, 199-205. Ogbobe, O. (1992). Physico-chemical composition and characterisation of the seed and seed oil of Sclerocarya birrea. Plant Food for Human Nutrition, 42, 201-206. Osman, M.A. (2004). Chemical and nutrient analysis of baobab (Adansonia digitata) fruit and seed protein solubility. Plant Food for Human Nutrition, 59, 29-33. Oyeleke, G.O., Salam, M.A., and Adetoro, R.O. (2012). Some aspects of nutrient analysis of seed, pulp and oil of baobab (Adansonia digitata L.). Journal of Environmental Science, Toxicology and Food Technology, 1, (4), 32-35. PhytoTrade Africa. (2008). Baobab (Adansonia digitata L.). Fruit pulp powder. Production capacity and sustainability in Southern Africa. Pretoria, South Africa, Report. Policy Briefing. (2007). Florestas e fauna bravia na segurança alimentar, nutrição e alívio à pobreza. www.researchgate.net, accessed at 2015-02-24, written in Portuguese.

6664 Prasanna, V., Prabha, T.N., and Tharanathan, R.N. (2007). Fruit ripening phenomena - an overview. Critical Reviews in Food Science and Nutrition, 47, 1-19. Proll, J., Petzke, K.J., Ezeagu, I.E., and Metges, C.C. (1998). Low nutritional quality of unconventional tropical crop seeds in rats - Nutrient requirements and interactions. Journal of Nutrition, 128, 2014-2022. Queface, A. (2009). Abordagem Geral Sobre desastres Naturais e Mudanças Climáticas em Moçambique. Instuituto Nacional de Gestão de calamidades Naturais em Colaboração com a Universidade eduardo Mondlane, report, http://desastres-moz.org/646656.pdf, accessed at 2012-09-05, written in Portuguese. Robinson, E., Lukman, A., and Bello, A. (2012). Investigation of extracted Sclerocarya birrea seed oil as a bioenergy resource for compression ignition engines. International Journal of Agriculture And Biology, 5(3), 60-67. Rodríguez R, Jiménez A, Fernández-Bolaños J, Guillén R, Heredia A (2006). Dietary fibre from vegetable products as source of functional ingredients. Trends in Food Science & Technology, 17(1), 3-15. Saka, J.D.K., and Msonthi, J.D. (1994). Nutritional value of edible fruits of indigenous wild trees in Malawi. Forest Ecology and Management, 64, 245-248. Saulawa, L.A., Yaradua, A.I., and Shuaibu, L. (2014). Effect of different processing methods on proximate, mineral and anti-nutritional factors content of baobab (Adansonia digitata) seeds. Pakistan Journal of Nutrition, 13 (6), 314-318. Schmidt, E., Lotter, M., and McCleland, M. (2002). Trees and shrubs of Mpumalanga and Kruger National Park. Jacana Media, Durban, South Africa. http://www.jacana.co.za/component/virtuemart/trees-and-shrubs-of-mpumulanga-and- kruger-national-park-detail?Itemid=0, accessed at 2015-05-02. Sena, L.P., VanderJagt, D.J., Rivera, C., Tsin, A.T.C., Muhamadu, I., Mahamadou, O., Millson, M., Pastuszyn, A., and Glew, R.H. (1998). Analysis of nutritional components of eight famine foods of the Republic of Niger. Plant Food for Human Nutrition, 52, 17-30. Sidibe, M. and Williams, J.T. (2002). Baobab. Adansonia digitata L. Ed. A. Hughes, N. Haq, R.W.Smith., University of Southampton, Southampton, SO17 1BJ, UK, http://zonedenial.com/documents/baobab%20Tree.pdf, accessed at 2012-02-27. Silva, G.G., Morais, P.L.D., Rocha, R.H.C., Santos, E.C., and Sarmento, J.D.A. (2009). Caracterização do fruto de cajaranazeira em diferentes estadios de maturação. Revista Brasileira de Produtos Agroindustriais, Campina Grande, 11, 159-163, written in Portuguese. Theander, O., Åman, P., Westerlund, E., Andersson, R., and Pettersson, D. (1995). Total dietary fiber determined as neutral sugar residues, uronic acid residues, and Klason lignin (the Uppsala method): collaborative study. Journal of AOAC International, 78, 1030-44.

6765 Tredgold, M. H., Biegel, H. M., Mavi, S., and Ashton, H. (1986). Food Plants of Zimbabwe, Mambo Press, 433, 1st edition, https://books.google.se/books/about/Food_Plants_of_Zimbabwe.html?id=bJA_AAAAYAAJ& hl=en, accessed at 2012-09-15. Uauy, R., Hoffman, D., Peirano, P., Birch, D., and Birch, E. (2001). Essential fatty acids in visual and brain development. Lipid, 36(9), 885-895.

UK Food Standards Agency. (2008). Baobab dried fruit pulp. http://acnfp.food.gov.uk/assess/fullapplics/baobab, accessed at 2012-11-07.

USDA, National Nutrient Database for Standard Reference Release 27. https://www.ars.usda.gov/main/site_main.htm?modecode=80-40-05-25, accessed at 2014-11- 22. Wairagu, N.W., Kiptoo, J., and Githiomi, J.K. (2013). Nutritional assessment of Sclerocarya birrea (Amarula) fruit from Kenya, International Journal of Current Research, 5 (5), 1074- 1078. Van Wyk, B., Van Wyk, P., and Van Wyk, B.E. (2008). Photo Guide to Trees of Southern Africa, Briza Publication, South Africa, 360 pages. Wehmeyer, A. S. (1966). The Nutrient composition of some edible wild fruits found in the Transvaal. South African Medical Journal, 40, 1102-1104. Whitney, E.N., and Rolfes, S.R. (1996). Understanding Nutrition. 7th edition, Minneapolis St. Paul, New York. WHO. (2007). Protein and amino acid requirements in human nutrition In: Report of a Joint WHO/FAO/UNU Expert Consultation, Agriculture Organization of the United Nations 935. http://whqlibdoc.who.int/trs/who_trs_935_eng.pdf, accessed at 2014-04-25. Wyk, B-E.V. (2005). Food Plants of the World Identification Culinary uses and Nutritional Values. Briza Publication, South Africa, 480 pages. Zharare, P., and Dhlamini, N. (2000). Characterization of marula (Sclerocarya caffra) kernel oil and assessment of its potential use in Zimbabwe. The Journal of Food Technology in Africa, 5(4), 126-128. Zimba, N., Wren, S., Stuck, A. (2005). Three major tree nut oils of southern central Africa: Their uses and future as commercial base oils. The International Journal of Aromatherapy, 15, 177-182. Websites: www.naturalhub.com accessed at 2015-04-18 www.gutsamba.co.mz accessed at 2015-04-18 www.cde.unibe.ch accessed at 2015-04-19 www.isptnature.org accessed at 2015-04-20

6866 Paper I

Hindawi Publishing Corporation The Scientific World Journal Volume 2013, Article ID 601435, 6 pages http://dx.doi.org/10.1155/2013/601435

Research Article Proximate Analysis of Five Wild Fruits of Mozambique

Telma Magaia,1,2 Amália Uamusse,3 Ingegerd Sjöholm,4 and Kerstin Skog2

1 Department of Biological Science, Science Faculty, Eduardo Mondlane University, Praca 25 de Junho, P.O. Box 257, Maputo, Mozambique 2 DivisionofAppliedNutritionandFoodChemistry,LundUniversity,P.O.Box124,22100Lund,Sweden 3 Department of Chemistry, Science Faculty, Eduardo Mondlane University, Praca 25 de Junho, P.O. Box 257, Maputo, Mozambique 4 Division of Food Technology, Lund University, P.O. Box 124, 221 00 Lund, Sweden

Correspondence should be addressed to Telma Magaia; [email protected]

Received 7 May 2013; Accepted 12 June 2013

Academic Editors: L. F. Goulao and L. Kitinoja

Copyright © 2013 Telma Magaia et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Mozambique is rich in wild fruit trees, most of which produce fleshy fruits commonly consumed in rural communities, especially during dry seasons. However, information on their content of macronutrients is scarce. Five wild fruit species (Adansonia digitata, Landolphia kirkii, Sclerocarya birrea, Salacia kraussii,andVangueria infausta) from different districts in Mozambique were selected for the study. The contents of dry matter, fat, protein, ash, sugars, pH, and titratable acidity were determined in the fruit pulps. Also kernels of A. digitata and S. birrea were included in the study. The protein content in the pulp was below 5 g/100 g of dry matter, but a daily intake of 100 g fresh wild fruits would provide up to 11% of the recommended daily intake for children from 4 to 8 years old. The sugar content varied between 2.3% and 14.4% fresh weight. The pH was below 3, exceptfor Salacia kraussii, for which it was slightly below 7. Kernels of A. digitata contained, on average, 39.2% protein and 38.0% fat, and S. birrea kernels 32.6% protein and 60.7% fat. The collection of nutritional information may serve as a basis for increased consumption and utilization.

1. Introduction regions [5–9], but the literature data on the nutritional value of wild fruits in Mozambique is limited [4, 10]. People in many InMozambique,alargenumberofwildfoodplantsarewidely communities are not aware of the nutritional value of the distributed throughout the country. The fruits and nuts are fruits; for example, they often eat only the pulp of the fruits sold at informal markets during the harvest season and are Sclerocarya birrea and Adansonia digitata while discarding consumed in various ways, and they are much appreciated the seeds, which contain a kernel with a higher protein and by children [1–3]. The importance of wild fruits in the diet fat content than peanuts [1, 11]. depends to a large extent on the availability of the fruits, since The aims of this work were to perform a study on cultivated fruit trees are not particularly common in the dry traditional utilization of wild fruits in Mozambique and regions of the country. Depending on the season, the fruits to generate data on the proximate composition and other are eaten raw, pressed for juice, cooked with sugar, or used characteristics of five wild fruits as a basis for the selection as flour to make porridge; the seeds or nuts are roasted to be of fruits suitable for processing. The long-term goal was to eaten as snacks. The choice of fruit species varies according promote and increase the utilisation and consumption of to region and cultural traditions [4]. indigenous fruits. The wild fruits selected for the present Many wild fruits and nuts are good sources of carbo- study were Adansonia digitata, Landolphia kirkii, Salacia hydrates, protein, fat, vitamins, and minerals that may be kraussii, Sclerocarya birrea, and Vangueria infausta.These deficient in common diets [5]. There are some reports on the fruits are popular in Mozambique, and they play an important chemical composition of wild fruits from Southern African role in the diet, particularly in rural areas. 2 The Scientific World Journal

2. Materials and Methods water, with 0.1 M sodium hydroxide using phenolphthalein as indicator [12]. All determinations were performed at least 2.1. Species. Five wild fruit species were studied: Adansonia in triplicate; the data are expressed as means ± standard digitata (A. digitata) (family Bombacaceae,localnamen’buyu deviations. Subsamples of fruits collected in 2009 were sent or Malambe), Landolphia kirkii (L. kirkii) (family Apocy- to an authorized laboratory for analysis of sugar content by naceae, local name n’vhungwa), Salacia kraussii (S. kraus- high performance anion exchange chromatography (Dionex) sii) (family Celastraceae, local name n’phinsha), Sclerocarya with pulsed amperometric detection (HPAEC-PAD). The birrea (S. birrea) (family Anacardiaceae,localnamen’canhi), variation between duplicate determinations was below 15%. and Vangueria infausta (V. infausta) (family Rubiaceae,local name n’pfilwa). Ripe fruits were collected in 2008 and 2009, 2.4. Statistical Analysis. Statistical analyses were performed except for the fruits from S. birrea,whichwerecollected using SPSS (version 13). Significant differences were evaluated only in 2009. A. digitata fruits, grown in the Tete district with one-way analysis of variance (ANOVA) followed by 1100kmfromMaputocity,wereboughtatalocalmarketin Tukey’s multiple comparisons test. A value 𝑃 < 0.05 was Maputo, and some fruits were collected in family orchards in considered to be significant. the Vilankulos district, 700 km south of Maputo. L. kirkii, S. kraussii, and V. infausta fruits were collected in orchards in the Marracuene and Manhic¸a districts, 30 and 50 km south 2.5. Interviews Regarding the Traditional Utilization of Wild of Maputo. Fruits from S. birrea were obtained from a garden Fruits. When the wild fruits were collected in the different in Maputo city and S. birrea kernels, dried for 1–3 months, districts, local people were interviewed about the traditional were obtained from a small family orchard in Manhic¸a. The useofthefruits.Foreachfruit,3–10peoplewereinterviewed, fruits were collected in districts where there is an increased both women and men of different ages, all of whom had occurrence and consumption of them. grown up in rural areas. The aim of the interviews was to obtain information on knowledge concerning the occurrence 2.2. Sample Preparation. Unblemished fruits were selected of these wild fruit trees, their owners, and traditional habits and washed, the skin and seeds were removed, and the regarding consumption, processing, and storage of wild remaining parts were homogenized in a blender to obtain fruits. 100 g pulp of each type of fruit. Different numbers of fruit Examples of the interview questions are as follows. (i) were used depending on fruit size and mass of pulp. The What kind of wild fruits exist in this area? (ii) Who owns fruits from A. digitata had low moisture content and the pulp and takes care of these fruit trees? (iii) Who are the major was ground into a fine powder and sieved (500 𝜇mmeshes). consumers? (iv) How are the fruits eaten? (v) What is the The seeds from A. digitata and S. birrea were crushed and typical amount harvested per day? (vi) How are the fruits the kernels inside were removed, milled, and sieved (500 𝜇m stored?(vii)Whatkindofprocessingcanbedone? meshes). Samples for determination of pH and titratable acidity were kept at room temperature and the analyses were 3. Results and Discussion performed on the day after collecting the fruit. The samples for the other analyses were vacuum-packed in plastic bags The results of the determinations of dry matter content ∘ andstoredat−18 C in a freezer. and proximate composition (protein, fat, and ash) for pulp andkernelsaresummarisedinTable1.Thevaluesdiffered 2.3. Analysis. To determine the dry matter content, 2 g somewhat between the years, but the difference was not ∘ samples were dried in an oven at 105 C until constant weight significant for any of the fruits. The dry matter content of A. [12]. The samples were weighed before and after drying digitata pulp was very high, on average 88.5%, while for the and the contents of dry matter were calculated. The protein other fruits, it varied between 16.7% and 34.8%. The high dry content was determined in an Elementar Analyzer (Flash EA matter content of A. digitata wasinthesamerangeashasbeen 1112 Series, Thermo Fisher Scientific, Sweden), by means of reported in studies in other countries, 85–95% [5, 7, 8, 14– combustion of 25 mg samples. Aspartic acid (Thermo Fisher 18]. There is little literature data on the dry matter content Scientific, Delft, The Netherlands) was used as a standard. of the other fruits, and for S. birrea our data agree with one The amount of protein was calculated by converting the report [7], while one report is showing a lower value [8]. For amount of nitrogen by a factor 6.25. The fat content was V. infausta ourresultsaresomewhathigherthanreported determined gravimetrically after extracting 1 g samples with in fruits from Malawi [5], Botswana [8], and Tanzania [17]. petroleum ether (Sigma-Aldrich Chemicals Co., St. Louis, For L. kirkii, the dry matter content was comparable with ∘ MO, USA) at 40–60 C for 1 hour using a Soxhlet equipment the literature data [1]. The average dry matter content of A. (SoxtecTM 2055, Foss, Hogan¨ as-Helsingborg,¨ Sweden) [13]; digitata kernelswas92.0%,whichisinagreementwithother rapeseed oil was used as a standard. The ash content was results, ranging from 85% to 97% [1, 7, 9, 14, 15, 18–21]. The determined by combustion of 2 g samples in silica crucibles average dry matter content of S. birrea kernels was 94.3%, in a muffle furnace (Carbolite, Sheffield, England) for24 whichisatthesamelevelastheresultsfromotherreports ∘ hours at 550 C[12]. The pH was determined using apH [1, 11, 15, 22, 23]. meter (Carison GLP 21, serial no. 147012, Barcelona, Spain). Theproteincontentofthepulpwaslowingeneral(below Titratable acidity, expressed in percentage of citric acid, was 5%) for all the fruits. This is in agreement with results of determined after titration of 10 g samples, dissolved in 100 mL other reports, [1, 5, 7, 15–18]. For A. digitata,however,there The Scientific World Journal 3

Table 1: Dry matter content, protein, fat, and ash expressed in g/100 g dry matter (𝑛=3).

Sample Dry matter (%) Protein Fat Ash Location year Adansonia digitata pulp Tete 2008 89.8 ± 0.1 2.4 ± 0.0 0.5 ± 0.1 5.5 ± 0.1 Tete 2009 89.1 ± 0.0 2.2 ± 0.1 0.7 ± 0.6 7.4 ± 0.0 Vilanculos 2009 86.5 ± 0.1 2.1 ± 0.0 0.5 ± 0.1 7.4 ± 0.0 Landolphia kirkii pulp Marracuene 2008 27.7 ± 0.2 2.1 ± 0.2 0.9 ± 0.1 2.9 ± 0.0 Marracuene 2009 23.9 ± 0.0 NA 0.4 ± 0.2 3.5 ± 0.0 Manhic¸a 2009 20.1 ± 0.3 1.7 ± 0.0 1.3 ± 0.1 3.0 ± 0.0 Salacia kraussii pulp Marracuene 2008 16.4 ± 0.3 3.7 ± 0.0 0.8 ± 0.5 4.8 ± 0.2 Manhic¸a 2008 17.3 ± 0.1 2.3 ± 0.0 1.5 ± 0.6 3.4 ± 0.0 Manhic¸a 2009 16.5 ± 0.1 2.1 ± 0.1 0.7 ± 0.2 3.6 ± 0.0 Sclerocarya birrea pulp Manhic¸a 2009 16.8 ± 0.1 1.4 ± 0.1 0.9 ± 0.3 3.0 ± 0.1 Vangueria infausta pulp Marracuene 2008 37.4 ± 0.5 2.9 ± 0.2 0.5 ± 0.8 7.8 ± 0.6 Marracuene 2009 30.0 ± 0.1 2.2 ± 0.3 0.7 ± 0.1 5.3 ± 0.0 Manhic¸a 2008 37.3 ± 0.9 4.7 ± 0.4 0.7 ± 0.2 5.7 ± 0.4 Manhic¸a 2009 34.5 ± 0.7 3.3 ± 0.8 0.2 ± 0.0 3.2 ± 0.2 Adansonia digitata kernel Tete 2008 93.6 ± 0.0 36.7 ± 0.9 35.0 ± 0.2 7.7 ± 0.0 Tete 2009 91.7 ± 0.1 38.6 ± 0.8 39.9 ± 6.2 7.2 ± 0.0 Vilanculos 2009 90.6 ± 0.0 42.7 ± 0.7 39.0 ± 7.0 8.5 ± 0.2 Sclerocarya birrea kernel Manhic¸a 2008 93.6 ± 0.2 30.1 ± 0.2 58.3 ± 1.3 3.8 ± 0.0 Manhic¸a 2009 95.0 ± 0.0 35.0 ± 0.1 63.1 ± 0.1 3.5 ± 0.0

is a report showing a much higher protein content, 15.3% been shown to contain high amounts of the essential amino [24]. The protein content in the pulp was significantly lower acids phenylalanine, lysine, and threonine [3, 6]. than in the corresponding kernels (𝑃 < 0.05). A. digitata The fat content in the pulp was below 2% for all the fruits. kernels contained on average 39.3% protein and S. birrea The literature data on A. digitata and S. birrea pulp generally kernels 32.6%. For A. digitata, our results are in accordance show fat contents below 1% [1, 7, 14–16, 18], while some with other results [7, 18],buttherearealsoreportsshowing reportsshowahigherfatcontentinA. digitata,4%[5, 17, 24]. lower protein content, 13–27% [14, 15, 19, 21], and one report The pulp of wild fruits is typically low in fat and protein [5], showing higher protein content, 48.3% [20]. The protein whilethekernelsaregoodsourcesoffatandprotein[10]. In content in the kernels of S. birrea was at the same level as the present study, the average fat content was 38.0% in kernels found in another report [23]. The high protein content in the of A. digitata and 60.7% in kernels of S. birrea,andthefat kernelsisatthesamelevelsasreportedforsoyabeans,around content in the kernels was significantly higher than that in the 33% [11], which means that the kernels may be a potential pulp (𝑃 < 0.05).Ourdataonkernelsareinagreementwith sourceofproteinandcanbeusedtoimprovethedietinrural results from other studies [1, 3, 22, 23], while the fat content communities.Forexample,adailyintakeof100goffresh was lower in two reports [20, 21]andhigherinone[18]. The pulp from the wild fruits studied here would provide around fat quality of the kernels is good according to the literature 2–11% of the recommended daily intake (RDI) for children data: A. digitata kernels are rich in palmitic, oleic, and linoleic from 4 to 8 years old, while 20 g of A. digitata or S. birrea acid (essential fatty acid) [15], and S. birrea kernels are rich in kernels would provide 32–39% of the RDI for children of the oleic and linoleic acid [3]. same age [25]. The protein quality of the kernels seems to be The average ash content ranged from 3.0% to 7.8%.For A. good since high amounts of the essential amino acid lysine as digitata pulpandkernel,theresultsareatthesamelevelasin well as of arginine, glutamic acid, and aspartic acid have been other reports [7, 8, 14–17, 21]. The ash content was somewhat reported for A. digitata kernels [15]. S. birrea kernels have lower than that in some other reports for S. birrea [7, 8, 23] 4 The Scientific World Journal and V. infausta pulp [5]. The high ash content indicates that Table 2: Sugar content of selected fruits obtained in 2009 from thefruitsandkernelsmaybegoodsourcesofminerals. Manhic¸a expressed in g sugar/100 g pulp (𝑛=2). The pH of the pulps showed an acidic character (around Sample Dry matter (%) Glucose Fructose Sucrose pH 3) except for S. kraussii, for which the pH was slightly above 6. The acidic character is in accordance with dataon Adansonia digitata 88.0 3.0 3.0 4.3 pulps from A. digitata, S. birrea, and V. infausta [8]. The Landolphia kirkii 18.0 7.5 5.7 1.2 pH of fruits generally varies between 2.5 and 4.5 due to Salacia kraussii 9.0 3.7 3.9 <0.1 their content of organic acids [26]; the low pH enhances the Sclerocarya birrea 8.0 0.5 0.4 1.4 microbiological and physicochemical stability [27]. Vangueria infausta 31.0 1.4 1.4 2.7 The titratable acidity of the pulp, which contributes to the acidity of the aroma, ranged from 0.6% to 1.7%. In another The variation between duplicate determinations was below 15%. report, also using citric acid, the titratable acidity was 7.8% for A. digitata, 0.9% for S. birrea, and 1.7% for V. infausta [8]. Comparable data were 0.3% for mango pulp [28]and0.7% Table 3: Traditional consumption and use of the studied fruits. for orange juice [29]. Table 2 shows the sugar content of the investigated fruits Species Utilisation expressed as g sugar/100 g pulp. The highest total sugar Pulp:fresh,diluted,andsweetenedtojuice, content was found in A. digitata and L. kirkii, 10.3 and Adansonia frozen to sweet ice, cooked to porridge, and digitata fermented to an alcoholic drink 14.4 g/100 g, respectively. The value for A. digitata is much lower than that reported in another study, where the total Kernels: fresh or roasted, milled and boiled to sugar content was around 30% [16]. The highest sucrose sauce or porridge Landolphia content, 4.3 g/100 g, was found in A. digitata, while for the Fresh or fermented to an alcoholic drink other fruits it was lower than 3 g/100 g. As expected, only kirkii Salacia very low amounts of maltose and lactose, below 0.04 g/100 g, Eaten fresh were detected; the most abundant sugars in fruits, are glucose, kraussii Pulp: fresh, squeezed to juice, and fermented to fructose, and sucrose in various proportions, depending on Sclerocarya an alcoholic drink species [30]. birrea Thesugarcontent,dataonpH,andtitratableacidityare Kernels: fresh or roasted, cooking oil can be essential characteristics, indicating the possibility for future extracted Fresh, soaked, squeezed to juice, mixed with use of these wild fruits. The sugar content is important for Vangueria sugar, water, or milk to a porridge, and fermented the development of the aroma and taste, and in product infausta development it is important to find a good balance between to an alcoholic drink pH, sugars, and titratable acidity to receive an optimal taste. The wild fruits in our study have different profiles regarding these characteristics but are in accordance with the literature data on some traditional fruits for juice production, for plastic bags and frozen. This sweet ice is commonly sold in example, papaya, mango, pineapple, and orange [29–32]. informal markets, and it is served as refreshment consumed by both children and adults. Interviews. The interviews revealed that the majority of the The seeds are crushed and the kernels inside can be fruits came from the forest and that wild fruits provide food consumed fresh or roasted. They can be milled to powder for everyone, especially for children because they are more and mixed with a small amount of water and boiled with local freetogointotheforesttocollectfruit.Somepeoplesaidthat plantfoodtomakeasauceconsumedwithboiledmaizeflour. in periods of hunger, the leaves, fruits, and seeds are used as The seeds are also boiled with a small amount of water or milk food as well as for medical applications. Wild fruits can serve to make porridge for children. as a source of income for many families, depending on the People in rural communities usually eat L. kirkii fruits amount of fruit that each family can reap in the forest or in the fresh, but when large amounts of these fruits are available, area surrounding their house. The fruit is usually sold early in they are squeezed and fermented to produce a local alcoholic the morning at informal markets to which people come to buy drinkthatisconsumedatsocialgatherings.Ruralpeopleofall fruit to sell in the markets in the cities. ages eat L. kirkii fruitwhilewalkingtoandworkingontheir Different fruits are used in different ways; see Table 3. A. grassland and cattle farm plots, which are sometimes far away digitata fruits are sold in different forms, for example, whole from their homes. fruit, pulp with seeds embedded, and fine powder made from The fruits from the small S. kraussii bushesaremoreeasily the pulp packed in plastic bags. The seed-containing pulp is accessible to children. Many school children eat the fresh often consumed fresh and soaked in warm water to remove fruits on their way to and from school or while grazing cattle. the seeds, and the remaining “milk” can be mixed with sugar Fresh fruit from S. birrea is squeezed to make juice or to form juice, or boiled with maize flour or sorghum to make fermented to produce a popular alcoholic drink. After juice a porridge given to children before they go to school. Another extraction and fermentation, the juice may be stored in sealed way of using the pulp is to dilute it with warm water to prepare clay pots or plastic containers for up to a year. The kernels ajuice,whichisfiltered,mixedwithsugarandpackedinsmall can be eaten fresh or roasted or ground in a mortar together The Scientific World Journal 5 with water, boiled with local plant food to make a sauce. The [5] J.D.KalengaSakaandJ.D.Msonthi,“Nutritionalvalueofedible kernels can also be used to produce oil for cooking. fruits of indigenous wild trees in Malawi,” Forest Ecology and In the southern part of Mozambique, V. infausta fruits Management,vol.64,no.2-3,pp.245–248,1994. are commonly consumed fresh, as juice, but they are also [6] R. H. Glew, D. J. Vanderjagt, C. Lockett et al., “Amino acid, often fermented to produce alcoholic drinks. Fresh fruit is fatty acid, and mineral composition of 24 indigenous plants of soaked in water, and the skin and seeds discarded before the burkina faso,” JournalofFoodCompositionandAnalysis,vol.10, preparation of a juice which is mixed with water and sugar or no.3,pp.205–217,1997. milk and served as porridge for children. Excess fruit is dried [7] S. S. Murray, M. J. Schoeninger, H. T. Bunn, T. R. Pickering, and andstoredforlateruse. J. A. Marlett, “Nutritional composition of some wild plant foods and honey used by hadza foragers of Tanzania,” JournalofFood Composition and Analysis,vol.14,no.1,pp.3–13,2001. 4. Conclusion [8] J. O. Amarteifio and M. O. Mosase, “The chemical composition of selected indigenous fruits of botswana,” Journal of Applied The wild fruits studied are consumed by people living in Science and Environmental Management,vol.10,no.2,pp.43– different districts of Mozambique and form a part of their 47, 2006. normal diet. Our data on the proximate composition, pH, [9] I.I.Nkafamiya,S.A.Osemeahon,D.Dahiru,andH.A.Umaru, titratable acidity, and sugar content are consistent with the “Studies on the chemical composition and physicochemical fewreportsavailableintheliterature.Weobservedlow properties of the seeds of baobab (Adasonia digitata),” African but not significant variations between the growth locations Journal of Biotechnology,vol.6,no.6,pp.756–759,2007. and the harvest year, and these variations may be due to [10] G. Saxon and C. Chidiamassamba, “Indigenous Knowledge of differences in climate, soil, and weather conditions. The Edible Tree Products—the Mungomu Tree in Central Mozam- findingsofthisstudyhaveshownthattheanalyzedfruits, bique. Links project. Gender biodiversity and local knowledge and especially the kernels, are good sources of protein and systems for food security,” FAO Report 40, 2005. fat. In Mozambique malnutrition is responsible for one-third [11] O. Ogbobe, “Physico-chemical composition and characterisa- of deaths in children under five years, and based on the above tion of the seed and seed oil of Sclerocarya birrea,” Plant Foods results, it may be concluded that promotion of consumption for Human Nutrition,vol.42,no.3,pp.201–206,1992. and processing of these fruits, to various products, may help [12] AOAC, OfficialMethodsofAnalysisoftheAssociationofOfficial to improve the diet and alleviate nutrient deficiencies. Analytical Chemists, 16th edition, 2000. [13] AOAC, OfficialMethodsofAnalysisoftheAssociationofOfficial Conflict of Interests Analytical Chemists, 17th edition, 1995. [14] C. T. Lockett, C. C. Calvert, and L. E. Grivetti, “Energy and The authors declare that they have no conflict of interests. micronutrient composition of dietary and medicinal wild plants consumed during drought. Study of rural Fulani, Northeastern Nigeria,” International Journal of Food Sciences and Nutrition, Acknowledgments vol.51,no.3,pp.195–208,2000. [15] M. A. Osman, “Chemical and nutrient analysis of baobab This study was financially supported by the Swedish Inter- (Adansonia digitata) fruit and seed protein solubility,” Plant national Development Agency Project “Technology Process- Foods for Human Nutrition,vol.59,no.1,pp.29–33,2004. ing of Natural Resources for Sustainable Development” at [16] A. A. Abdalla, M. A. Mohammed, and H. A. Mudawi, “Pro- Eduardo Mondlane University, Maputo. The authors thank duction and quality assessment of instant baobab (Adansonia Mrs. Lisbeth Person and Mr. Dan Johnson for excellent digitata L.),” Advance Journal of Food Science and Technology, technical assistance. vol. 2, no. 2, pp. 125–133, 2010. [17]T.V.Emmanuel,J.T.Njoka,L.W.Catherine,andH.V. References M. Lyaruu, “Nutritive and anti-nutritive qualities of mostly preferred edible woody plants in selected drylands of Iringa [1] P. M. Maundu, G. W. Ngugi, and C. H. S. Kakuye, Traditional District, Tanzania,” Pakistan Journal of Nutrition,vol.10,no.8, Food Plants of Kenya, National Museums, Nairobi, Kenya, 1999. pp.786–791,2011. [2] L. E. Grivetti and B. M. Ogle, “Value of traditional foods [18]A.E.Assogbadjo,F.J.Chadare,R.G.Kaka¨ı, B. Fandohan, and in meeting macro- and micronutrient needs: the wild plant J. J. Baidu-Forson, “Variation in biochemical composition of connection,” Nutrition Research Reviews,vol.13,no.1,pp.31– baobab (Adansonia digitata) pulp, leaves and seeds in relation 46, 2000. to soil types and tree provenances,” Agriculture, Ecosystems and [3] R. S. Glew, D. J. VanderJagt, Y.-S. Huang, L.-T. Chuang, R. Environment,vol.157,pp.94–99,2012. Bosse, and R. H. Glew, “Nutritional analysis of the edible pit [19]L.J.Proll,K.J.Petzke,I.E.Ezeagu,andC.C.Metges,“Low of Sclerocarya birrea in the Republic of Niger (daniya, Hausa),” nutritional quality of unconventional tropical crop seeds in Journal of Food Composition and Analysis,vol.17,no.1,pp.99– rats,” The Journal of Nutrition,vol.128,no.11,pp.2014–2022, 111, 2004. 1998. [4]P.D.MangueandM.N.Oreste,Data Collection and Anal- [20]H.O.Adubiaro,O.Olaofe,E.T.Akintayo,andO.-O.Babalola, ysis for Sustainable Forest Management in ACP Countries— “Chemical composition, cacium, zinc and phytate interrela- Linking National and International Efforts, EC-FAO Partnership tionships in baobab (Adansonia digitata) seed flour,” Advance Programme (1998–2000), Country Brief on Non-Wood Forest Journal of Food Science and Technology,vol.3,no.4,pp.228– Products, Maputo, Mozambique, 1999. 232, 2011. 6 The Scientific World Journal

[21] C. Parkouda, H. Sanou, A. Tougiani et al., “Variability of Baobab (Adansonia digitata L.) fruits’ physical characteristics and nutrient content in the West African Sahel,” Agroforestry Systems, vol. 85, no. 3, pp. 455–463, 2011. [22] I. C. Eromosele and C. O. Eromosele, “Studies on the chemical composition and physico-chemical properties of seeds of some wild plants,” Plant Foods for Human Nutrition,vol.43,no.3,pp. 251–258, 1993. [23]S.Muhammad,L.G.Hassan,S.M.Dangoggo,S.W.Hassan,K. J. Umar, and R. U. Aliyu, “Nutritional and antinutritional com- position of Sclerocarya birrea seed kernels,” Studia Universitatis “Vasile Goldis¸”, Seria S¸tiint¸ele Viet¸ii,vol.21,no.4,pp.693–699, 2011. [24] I. C. Obizoba and N. A. Amaechi, “The effect of processing methodsonthechemicalcompositionofbaobab(Adansonia digitata L.),” Pulp and Seed. Ecology of Food and Nutrition,vol. 29, pp. 199–205, 1993. [25] The National Academies of Science, Dietary Reference intakes for Energy, Carbohydrates, Fiber, Fat, fatty acids, (Macronutri- ents), 2005, http://www.nap.edu/. [26] A. L. Calvacanti, K. F. Oliveira, P. S. Paiva, M. V. Rebelo Dias, S. K. P. Costa, and F. F. Vieira, “Determination of total soluble solids content (Brix) and pH in milk drinks and industrialized fruit juices,” Pesquisa Brasileira em Odontopediatria e Cl´ınica Integrada,vol.6,no.1,pp.57–64,2006. [27] P. R. Gayon, Y. Glories, A. Maujean, and D. Dubourdieu, “Organic acids in wine Handbook of Enology: the Chemistry of Wine and Stabilization and Treatments,”vol. 2, pp. 9–10, John Wiley & Sons, England, UK, 2nd edition, 2006. [28] N. I. Singh, C. Dhuique-Mayer, and Y. Lozano, “Physico- chemical changes during enzymatic liquefaction of mango pulp (cv. Keitt),” Journal of Food Processing and Preservation,vol.24, no. 1, pp. 73–85, 2000. [29] B. K. Tiwari, K. Muthukumarappan, C. P. O’Donnell, and P. J. Cullen, “Kinetics of freshly squeezed orange juice quality changes during ozone processing,” Journal of Agricultural and Food Chemistry, vol. 56, no. 15, pp. 6416–6422, 2008. [30] F. J. Rambla, S. Garrigues, and M. De La Guardia, “PLS-NIR determination of total sugar, glucose, fructose and sucrose in aqueous solutions of fruit juices,” Analytica Chimica Acta,vol. 344, no. 1-2, pp. 41–53, 1997. [31] R. Shamsudin, W. R. W. Daud, M. S. Takriff, and O. Hassan, “Physicochemical properties of the Josapine variety of pineap- ple fruit,” International Journal of Food Engineering,vol.3,no.5, article 9, 2007. [32] L. E. Gayosso-Garc´ıaSancho,E.M.Yahia,M.A.Mart´ınez- Tellez,´ and G. A. Gonzalez-Aguilar,´ “Effect of maturity stage of papaya maradol on physiological and biochemical parameters,” American Journal of Agricultural and Biological Science,vol.5, no. 2, pp. 194–203, 2010. Paper II

Magaia et al. SpringerPlus 2013, 2:88 http://www.springerplus.com/content/2/1/88 a SpringerOpen Journal

RESEARCH Open Access Dietary fiber, organic acids and minerals in selected wild edible fruits of Mozambique Telma Magaia1,3*, Amália Uamusse2, Ingegerd Sjöholm3 and Kerstin Skog3

Abstract The harvesting, utilization and marketing of indigenous fruits and nuts have been central to the livelihoods of the majority of rural communities in African countries. In this study we report on the content of dietary fiber, minerals and selected organic acids in the pulps and kernels of the wild fruits most commonly consumed in southern Mozambique. The content of soluble fiber in the pulps ranged from 4.3 to 65.6 g/100 g and insoluble fiber from 2.6 to 45.8 g/100 g. In the kernels the content of soluble fiber ranged from 8.4 to 42.6 g/100 g and insoluble fiber from 14.7 to 20.9 g/100 g. Citric acid was found in all fruits up to 25.7 g/kg. The kernels of Adansonia digitata and Sclerocarya birrea were shown to be rich in calcium, iron, magnesium and zinc. The data may be useful in selecting wild fruit species appropriate for incorporation into diets. Keywords: Wild fruits, Minerals, Citric acid, Dietary fiber, Daily intake

Background digitata), as a novel food ingredient (Commission Deci- Fruits are generally recognized as essential for health sion 2008). Three components are of special importance optimisation, with human health depending to a large for determining whether such ingredients are health- extent on factors such as high fruit and vegetable con- enhancing: the kind and amount of minerals, organic acids sumption (Ibrahim 2011). Deficiencies of essential mi- and dietary fiber. cronutrients found in fruits can increase the risk of Minerals are of great importance in the diet, although illness or death from infectious diseases by reducing im- they comprise only 4–6% of human bodyweight. Some mune and non-immune defenses and by compromising minerals or macro elements required in amounts greater normal physiology and development (Black 2003). Such than 100 mg per day represent 1% or less of bodyweight nutrient deficiencies are highly prevalent in low and (Insel et al. 2011; Imelouane et al. 2011). The essen- middle income countries. tial macro elements include calcium, phosphors, magne- Recent research has shown that a wide range of indi- sium, potassium, sodium, sulfur and chloride. Essential genous fruit trees have the potential to provide rural trace elements such as zinc, iron, copper, manganese, households with a means to meet their nutritional and selenium, iodine and molybdenum are normally required medicinal needs (Ekesa et al. 2009). In the past decades in amounts of less than 100 mg per day, making up less several reports have been published on the nutritional than 0.01% of the bodyweight (Imelouane et al. 2011). composition of wild fruits and vegetables growing in dif- Dietary fiber is increasingly viewed as an essential as- ferent areas in various African countries and on the ef- pect of good nutrition. Intake of dietary fiber alters the fect they could have on combating malnutrition and water content, viscosity, and microbial mass of the intes- poverty in the continent (FAO 2011). In June 2008, the tinal contents, resulting in changes in the rate and ease European Commission authorised the placing on the mar- of passage through the intestine (Elleuch et al. 2011). ket of dried pulp of one wild fruit, Baobab (Adansonia High intake of dietary fiber plays a significant role in weight control and the prevention of several diseases. * Correspondence: [email protected] For example, dietary fiber improves glucose tolerance, 1 Department of Biological Science, Science Faculty, Eduardo Mondlane by delaying the transport of carbohydrates into the small University, PO Box 257, Maputo, Mozambique 3Department of Food Technology, Engineering and Nutrition, Lund intestine, reducing the risk of heart diseases and reduces University, PO Box 124, Lund SE-221 00, Sweden constipation (Anderson et al. 1994; Rodríguez et al. Full list of author information is available at the end of the article

© 2013 Magaia et al.; licensee Springer. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Magaia et al. SpringerPlus 2013, 2:88 Page 2 of 8 http://www.springerplus.com/content/2/1/88

2006). Organic acids are involved in human growth, these fruits. The dietary fiber, organic acids and mineral maturation and senescence (Al-Farsi et al. 2005). The or- content of the five wild fruits and selected seeds were ganic acids influence organoleptic properties such as fla- determined with the plan to highlight their potential as vor, color and aroma and are responsible for many an effective means to combat macro and micronutrient characteristic fruity tastes. They increase shelf life, stabil- deficiencies, especially in children. ity and microbiologic safety (Hasib et al. 2002; Loredana et al. 2006; Nour et al. 2010). Results In a previous study we reported on the proximate The contents of insoluble and soluble dietary fiber in the composition of five wild fruits of Mozambique. Adan- fruits are presented in Table 1. All pulps and kernels sonia digitata, Landolphia kirkii, Salacia kraussii, Scle- contained dietary fiber, but with large variations in con- rocarya birrea and Vangueria infausta were evaluated centration among the different fruits. In S. kraussii and for pH and titratable acidity and their content of dry in V. infausta, the content of insoluble dietary fiber matter, fat, protein, ash, soluble solids and sugar content ranged from 2.6 g/100 g to 45.8 g/100 g in the pulps, (Magaia et al. submitted). The present study was car- and from 14.7 to 20.9 g/100 g in the kernels, respect- ried out to further analyze the nutritional potential of ively. The content of insoluble dietary fiber in A. digitata

Table 1 Insoluble and soluble dietary fiber content expressed on the basis of dry matter (n=3)* Fruit part, Location_year Dry matter Insoluble dietary fiber (g/100 g) Soluble dietary fiber Adansonia digitata pulp Tete_2008 86.4±0.5 14.2±0.4 60.3±1.9 Tete_2009 86.5±0.1 14.7±0.0 65.6±0.6 Vilankulos_2009 85.6±0.5 16.1±0.8 57.3±0.3 Adansonia digitata kernels Tete_2008 92.7±0.3 17.4±3.4 14.0±0.6 Tete_2009 91.7±0.1 20.9±1.3 17.0±3.4 Vilankulos_2009 90.1±0.1 14.7±0.0 42.6±1.8 Adansonia digitata whole seed Tete_2009 98.2±0.0 56.5±3.1 15.9±1.5 Vilankulos_2009 96.4±0.0 62.0±1.2 16.3±0.6 Landolphia kirkii pulp Marracuene_2008 26.9±0.1 3.5±0.6 4.6±0.1 Marracuene_2009 23.9±0.2 4.9±0.8 4.3±0.3 Manhica_2009 20.4±0.2 4.9±0.4 5.8±1.2 Salacia kraussii pulp Marracuene_2008 16.1±037 3.3±0.1 6.0±0.4 Manhica_2008 14.0±0.7 2.6±1.2 7.6±1.0 Manhica_2009 16.5±0.2 5.5±0.3 7.1±0.4 Sclerocarya birrea pulp Manhica 2009 16.7±0.0 7.7±1.7 10.5±0.9 Sclerocarya birrea kernels Manhiça 2008 93.6±0.2 17.6±2.9 10.5±2.7 Manhiça 2009 95.0±0.0 18.5±1.5 8.4±2.4 Vangueria infausta pulp Marracuene_2008 34.9±0.2 45.6±1.4 24.3±1.4 Marracuene_2009 30.0±0.1 45.8±0.9 26.3±0.9 Manhica_2008 27.9±0.2 41.0±0.8 10.6±0.8 Manhica_2009 34.5±0.7 30.9±1.9 23.1±1.9 * ± indicates standard deviation. Magaia et al. SpringerPlus 2013, 2:88 Page 3 of 8 http://www.springerplus.com/content/2/1/88

pulp was significantly higher than that in L. kirkii and S. Discussion kraussii. The soluble dietary fiber in the pulps ranged The literature provides little data on dietary fiber in the from 4.3 g/100 g in L. kirkii up to 65.6 g/100 g in A. fruits investigated and we have found no data on soluble digitata. The content of soluble dietary fiber in A. or insoluble dietary fiber. Our data thus contributes sub- digitata pulp was significantly higher than in the other stantially to knowledge about dietary fiber content in fruit pulps. wild fruits. The analyses conducted showed that all fruits A chromatogram from the analysis of organic acids in selected for the study contained both soluble and insol- S. kraussii is shown in Figure 1. The peaks correspond- uble dietary fiber, and the total amount of dietary fiber ing to the different acids are indicated by numbers. The in the pulps ranged from around 10 to 80 g/100 g dry calibration curve was linear within the concentration matter (Table 1), thus indicating their potential for im- ranges used for the analysis. proving nutritional content in the local diet. The results of the determination of organic acid con- The largest amount of dietary fiber was found in tent based on wet weight are shown in Table 2. Fruits A. digitata pulp. The amount 80.3 g/100 g was much collected in Manhiça, 2009 were used for this analysis. higher than in previous studies, which ranged from Citric acid was found in all fruits, with the highest con- 5.4 g/100 g in fresh weight to 45 g/100 g in dry weight tents in A. digitata (25.7 g/kg) and L. kirkii (21.5 g/kg). (Saka and Msonthi 1994; Lockett et al. 2000; Murray Malic acid was detected at concentrations between et al. 2001; Osman 2004). The high content of dietary fi- 0.4 g/kg and 2.1 g/kg. Succinic acid was found at concen- ber in the present study, of which 80% was soluble diet- trations of 0.1 g/kg or below. Tartaric acid was detected at ary fiber, could thus provide a substantial contribution trace levels only in S. birrea. to total dietary intake. For example, the consumption of The mineral contents in the fruit pulps, kernels and 20 g A. digitata pulp can supply 42 – 52% of the re- seeds, expressed as mg/100 g dry matter, are presented commended daily intake (RDI) for children from 4 to in Table 3. The relative standard deviation of the mean 13 years of age and also for pregnant women (National values was generally below 5%. The content in different Research Council 2005). The high content of soluble fruit samples varied considerably, from undetectable dietary fiber could also help to control serum cholesterol levels of selenium up to 2753 mg/100 g for potassium levels, reduce other risk factors for cardiovascular dis- (S. birrea pulp). The richest source of calcium was A. ease (Van Duyn and Pivonka 2000), and reduce appetite digitata both for pulp (308 to 366 mg/100 g) and for and caloric intake (Anderson et al. 1994). kernels (293 to 347 mg/100 g.) The iron content in the V. infausta pulp also contained large amounts of total pulps and kernels generally ranged from 1.0 - 4.0 mg/ dietary fiber (on average 62 g/100 g), of which more 100 g, while S. kraussii pulp contained 9.0 mg/100 g, than half was insoluble dietary fiber. Several studies have and the whole seeds of A. digitata contained 29 mg /100 g. shown that intake of fruits rich in insoluble dietary fiber The highest content of magnesium was found in the ker- benefits weight control and health of the large intestine; nels of A. digitata (626 to 706 mg/100 g) and S. birrea the insoluble fiber decreases intestinal transit time and (396 – 436 mg/100 g). Selenium was not analyzed in all increases fecal bulk (Al-Farsi et al. 2005; Hamilton et al. fruits, but was detected in V. infausta (1.5 mg/100 g). 1992). Consumption of 100 g V. infausta pulp can sup- The richest source of zinc was the kernels of A. digitata ply up to 40% of RDI for children and pregnant women (5.2 – 5.7 mg/100 g) and S. birrea (4.5 mg/100 g). (National Research Council 2005). In a report on die-

Figure 1 Chromatogram from the analysis of organic acids in Salacia kraussii. The peaks correspond to 1- citric acid, 2- tartaric acid, 3- malic acid, 4- succinic acid, 5 and 6 unknowns. Magaia et al. SpringerPlus 2013, 2:88 Page 4 of 8 http://www.springerplus.com/content/2/1/88

Table 2 Organic acid content in fruit pulp expressed on the basis of wet weight (n = 3)* Sample Citric acid Tartaric acid Malic acid (g/kg) Succininc acid Adansonia digitata 25.7±2.1* ND 1.6±0.2 0.1±0.0 Landolphia kirkii 21.5±2.1 ND 0.4±0.1 0.03±0.0 Salacia kraussii 0.9±0.3 ND 1.1±0.1 0.1±0.0 Sclerocarya birrea 8.5±1.3 trace 1.2±0.2 0.1±0.0 Vangueria infausta 6.2±0.5 ND 2.1±0.2 0.1±0.0 * ± indicates standard deviation, ND= not detected. tary fiber in V. infausta, the content of acid detergent The total amount of dietary fiber in L. kirkii pulp was lignin was reported to be 35.5% (fresh weight), acid de- low (10 g/100 g) compared with that of the other fruits tergent fiber 39.5% and neutral detergent fiber 39.4% in this study, but very similar to that of L. oweriensis,of (Amarteifio and Mosase 2006). The discrepancies bet- the same family (Effiong and Udo 2010). The amount of ween our findings and previous research may be due to the dietary fiber in S. kraussii pulp was similar to L. kirkii. difficulty in comparing data on dietary fiber when it isn’t The amounts of dietary fiber in S. krausii, and L. kirkii clear whether figures refer to dry or fresh weight. Further- pulps are in the same range as in for example avocado, more, different analytical methods may have been used in 6.7 g/100 g, and guava, 12.7 g/100 g (Li et al. 2002). different studies, including different types of dietary fiber. The whole seeds of A. digitata contained large Our data on the amounts of dietary fiber in S. birrea amounts of insoluble dietary fiber. These seeds are often pulp were about half of that found in another study de- crushed into a powder in rural cooking and mixed with termining dietary fiber after extraction of fat by a gravi- other ingredients to make a sauce. Increased use of this metric method, 37.7 g/100 g (Murray et al. 2001). In a local custom may thus help to increase the intake of report where AOAC method was used, the amount of dietary fiber. acid detergent lignin was 13.7% (fresh weight), acid de- In the kernels of A. digitata the total content of diet- tergent fiber 16.3% and neutral detergent fiber 16.1% ary fiber was on average 42.2 g/100 g, which is higher (Amarteifio and Mosase 2006). than that found in other studies (Murray et al. 2001).

Table 3 Mineral content in the pulps, kernels and whole seeds of the studied wild fruits in Mozambique Fruit part, Location_year Dry matter g/100 g Ca K Fe Mg Na mg/100 g P S Se Zn Adansonia digitata pulp Tete_2008 90.1 326 2308 2.0 162 5.0 40 51 nd - Tete_2009 86.4 308 2392 2.0 129 2.0 40 41 - 0.5 Vilankulos_2009 86.5 366 2360 1.0 102 5.0 30 43 - 0.6 Adansonia digitata kernels Tete_2009 92.2 293 1416 6.0 626 1.0 1229 263 - 5.7 Vilankulos_2009 90.6 347 1451 4.0 706 2.0 1518 269 - 5.2 Adansonia digitata whole seed Tete_2008 93.6 220 1238 29 360 2.0 523 131 nd - Landolphia kirkii pulp Manhiça 2009 20.4 28 1840 4.0 51 21 55 20 - 1.4 Salacia kraussii pulp Manhiça 2009 8.9 127 2056 9.0 207 35 153 191 - 0.6 Sclerocarya birrea pulp Manhiça 2009 15.9 201 2753 3.0 138 30 178 90 - .0.7 Sclerocarya birrea kernels Manhiça 2008 92.7 60 622 4.0 436 7.0 871 284 nd - Manhiça 2009 95.0 81 531 4.0 396 6.0 719 269 - 4.5 Vangueria infausta pulp Manhiça 2009 34.4 90 1249 3.0 65 18 92 45 2.0 - The relative standard deviation was in general < 5%, nd = Not detectable; - = not analyzed. Magaia et al. SpringerPlus 2013, 2:88 Page 5 of 8 http://www.springerplus.com/content/2/1/88

The kernels of S. birrea contained similar amounts of in- with phosphor, magnesium and potassium is important soluble dietary fiber as A. digitata kernels, but the for growth and maintenance of bone, teeth and muscle amount of soluble dietary fiber was lower. There is no (Insel et al. 2011) and bone metabolism (Ilich et al. 2003; report in the literature about dietary fiber of S. birrea New 2003; Bonjour et al. 2009). kernels. However, in cashew nuts and peanuts, which High amounts of phosphor were found in the kernels: are commonly consumed Mozambique, the total dietary A. digitata kernels contained up to 1500 mg/100 g, which fiber content was 3.9 g/100 g for cashew nuts and 5.2 g/ is four times higher than that in another study (Nnam and 100 g for peanuts in fresh weight (de Oliveira Sousa Obiakor 2003). Regarding the pulp, there are reports et al. 2011). showing higher phosphor content than in our study, for The effects of dietary fiber cannot, however, be consid- example 452 mg/100 g in A. digitata (Sena et al. 1998) ered in isolation. Although dietary fiber provides many and 264 mg/100 g in S. birrea (Glew et al. 1997). health benefits, it may affect mineral absorption nega- High content of potassium, more than 2000 mg/100 g, tively due to the capacity of the fibers to bind cations was found in pulps from A. digitata, S. kraussii and S. (López and Martos 2004). The ability of dietary fiber birrea. For A. digitata, this is in agreement with results to interfere with iron absorption is especially negative from other studies (Saka and Msonthi 1994; Amarteifio for human nutrition (Reinhold et al. 1981). However, cit- and Mosase 2006), while some reports show lower ric and malic acids in fruits promote iron absorption values. For S. birrea, our data agree with other results (Gillooly et al. 1983), and improve iron solubilization (Amarteifio and Mosase 2006). The content of potas- (López and Martos 2004). sium in V. infausta pulp was on the same level as in one Fortunately, citric and malic acid were found in all report (Amarteifio and Mosase 2006) but almost 7 times fruits in this study (Table 2). Citric acid ranged from greater than in another report (Saka and Msonthi 1994). 0.9 g/kg of wet weight basis in S. Kraussii to more than The potassium content in kernels of S. birrea was some- 20 g/kg in A. digitata and L. kirkii. For malic acid the re- what higher than that in another report (Glew et al. sults were lower and ranged from 0.4 g/kg to 2.1 g/kg. 2004). In the whole seeds of A. digitata, potassium is Succinic acid was found in low amounts in all fruits, higher than in other reports (Osman 2004; Nkafamiya around 0.1 g/kg, and tartaric acid was detected in trace et al. 2007). Potassium, together with sodium, regulates levels in S. birrea. No data in the literature are available muscle contraction and nerve impulse transmission, and on organic acids in the selected wild fruits; however, one a high potassium/sodium ratio may assist the excretion type of wild fruit called medlar (Mespilus germanica L.) of excessive salt and water (Arthey et al. 2001). All fruits was reported to contain around 4 g/kg fresh weight of in our study had low amounts of sodium and thus the citric acid and malic acid (Glew et al. 2003). For com- potassium/sodium ratio was high in the fruits. parison, the citric acid concentrations in some trad- The whole seeds of A. digitata had extremely high itional fruits are: pineapple (2.2 g/kg), orange (4.5 g/kg), iron content, 29 mg/100 g, while data in the literature grapes (13.1 g/kg) and lime (41.2 g/kg) (Falade et al. range from 1.83 to 6.36 mg /100 g (Lockett et al. 2000; 2003). Osman 2004; Glew et al. 1997; Nkafamiya et al. 2007). S. Although there were large variations in the mineral kraussii had the highest iron content of the pulp, 9 mg/ content in the different fruits (Table 3), we did not ob- 100 g. In the other pulps it was 2 to 4 mg/100 g. In the serve any pronounced difference between growth place literature, the amounts of iron in A. digitata ranged or harvest year, and the small differences found may be from 0.1 to 9.3 mg/100 g (Saka and Msonthi 1994; explained by soil, climate and weather conditions. Lockett et al. 2000; Amarteifio and Mosase 2006; Glew The highest amounts of calcium from the fruits se- et al. 1997; Eromosele et al. 1991; Sena et al. 1998; Glew lected in the present study were found in A. digitata et al. 2004) and in S. birrea from 0.07 to 2.49 mg/ pulp and kernel, around 300 mg/100 g dry matter. The 100 g (Amarteifio and Mosase 2006; Glew et al. 1997; amounts are at the same levels as in other reports Eromosele et al. 1991). For V. infausta there is one re- (Osman 2004; Glew et al. 1997) but higher than in some port showing 0.09 mg /100 g (Amarteifio and Mosase studies (Saka and Msonthi 1994; Lockett et al. 2000; 2006) and one showing 28.3 mg/100 g (Saka and Amarteifio and Mosase 2006; Eromosele et al. 1991; Msonthi 1994). Very high iron content, 129 mg/100 g Sena et al. 1998). S. birrea pulp contained 201 mg/100 g, dry matter, was reported in L. oweriensis, which is a fruit which is about half the amount reported in the literature in the same family as L. kirkii (Effiong and Udo 2010). (Glew et al. 1997), but higher than that in other reports Iron is necessary for the transport of oxygen in the (Amarteifio and Mosase 2006; Eromosele et al. 1991). blood to the cells and for supplying the body with Calcium is an important factor in bone health, and a energy, for immune function and nerve health and in high intake is recommended particularly during preg- addition, it is a cofactor in numerous reactions (Insel nancy and infancy (Insel et al. 2011) calcium together et al. 2011). Magaia et al. SpringerPlus 2013, 2:88 Page 6 of 8 http://www.springerplus.com/content/2/1/88

The kernels of A. digitata had the highest magnesium to income generation and stimulating rural economic content, around 600 – 700 mg/100 g. This is higher than development. the magnesium content in other kernels or seeds. For example dried pumpkin seeds contain 540 mg/100 g, lin- Materials and methods seed 392 mg/100 g and sunflower seeds 355 mg/100 g Fruit samples (National Food Agency 2012). Our data on whole seeds Five fruits were used for the analysis: Adansonia digitata are at the same level as in other reports for A. digitata (A. digitata) (Fam. Bombaceae, local name n‘buvu or (Lockett et al. 2000; Osman 2004; Glew et al. 1997) and malambe), Landolphia kirkii (L.kirkii) (Fam. Apocyna- for S. birrea (Glew et al. 2004). Magnesium is of great ceae, local name wungwa), Salacia kraussii (S. kraussii) importance for cardiac and nerve function, is involved in (Fam. Celastraceae, local name phinsha), Sclerocarya more than 300 biochemical reactions in the body and is birrea (S. birrea) (Fam. Anacardeaceae, local name canhi) involved in energy metabolism and protein synthesis and Vangueria infausta (V. infausta) (Fam. Rubiaceae, (Insel et al. 2011). local name pfilwa). Approximately 5 kg of each fruits were The zinc content in the kernels of A. digitata and S. collected in January to July in 2008 and 2009 in four dis- birrea was around 5 mg/100 g, which is in agreement tricts of Mozambique, with the exception of the fruits with literature for S. birrea (Glew et al. 2004) and higher from Sclerocarya birrea, which were collected only in than previous data for A. digitata (Nnam and Obiakor 2009. Unblemished fruits were selected and washed, the 2003). Zinc increases the affinity of hemoglobin for oxy- skin and seeds were removed and the remaining parts gen, participates in taste perception and interacts with a (pulp) were homogenized in a blender to obtain 100 g number of hormones. In addition, the body needs zinc pulp of each type of fruit. Different numbers of fruit were to grow and develop properly during pregnancy, infancy, used, depending on size and mass of pulp. Seeds from and childhood (Brown et al. 1999; Insel et al. 2011). Adansonia digitata and Sclerocarya birrea were crushed The results of the mineral analysis can be compared and the kernels were removed, milled and sieved. The dry with the RDI for children 4–13 years of age and preg- matter content in the pulps was determined immediately. nant women 19–30 years of age (National Research Fruit pulps and kernels for analysis of dietary fibre, or- Council 2005). For example, 100 g of fresh A. digitata ganic acids and minerals were vacuum packed in plastic pulp can contribute on average 23% of the iron and 30% bags and stored at −18°C. of the calcium RDI for children (4–13 years) and almost 29% of the calcium for pregnant women. Furthermore, Chemicals 100 g S. birrea pulp can contribute 13% of the magne- Chemicals and solvents were of analytical grade. Sulphuric sium RDI for children (4–13 years) and 33% for preg- acid, nitric acid, hydrochloric acid, sodium di-hydrogen nant women, and almost 41% of zinc requirements. phosphate, di-sodium hydrogen phosphate; sodium hy- Consumption of 60 g A. digitata kernels can supply droxide, ethanol and acetone were from Fluka (Sigma- around 30% of iron, almost 50% of zinc and more than Aldrich, Steinheim, Germany). Pepsin (2000FIP U/g) 100% of magnesium RDI for children. obtained from Merck (Darmstadt, Germany), pancreatin from Fluka (Sigma-Aldrich, Steinheim, Germany), and Conclusion celite were used for the analysis of dietary fibre. Lithium New data on dietary fiber, organic acids and mineral chloride, organic acid standards (citric, malic, tartaric and content have been obtained for five wild fruits and se- succinic) were from Merck (Darmstadt, Germany). lected seeds. The highest content of dietary fiber was found in A. digitata pulp. It was also found that fresh A. Analysis digitata pulp can contribute a large amount of the iron Dry matter and calcium RDI for children and that kernel of A. To determine the dry matter content, 2 g samples were digitata and S. birrea can contribute significantly to the dried in an oven at 105°C until constant weight (AOAC magnesium and zinc requirements for pregnant women. 2000 method 920.151). The samples were weighed be- Thus some of the fruits and kernels studied show large fore and after drying and the contents of dry matter potential to reduce mineral deficiencies in local diet es- were calculated. All determinations were performed in pecially in children. The research results highlight the triplicate. significance of wild edible fruits as a cheap source of nu- trients and the benefits of increasing the use these Dietary fiber species as dietary supplements. Initiatives should be The content of total dietary fiber was first analyzed using put in place to promote consumption and domestica- an enzymatic gravimetric method, and then divided into tion of edible indigenous fruit: to improve the nutritional fractions of either soluble dietary fiber (SDF) or insoluble and health status of women and children, contributing dietary fiber (IDF) (Asp et al. 1983). All experiments were Magaia et al. SpringerPlus 2013, 2:88 Page 7 of 8 http://www.springerplus.com/content/2/1/88

performed in triplicate. The sample (0.5-1.0 g) was Statistical analyses suspended in a phosphate buffer and hydrochloric acid The results of the dietary fibre were subjected to analysis was added to adjust the pH to 1.5. The sample was of variance. Differences between means were tested at digested by pepsin for 60 min at +4°C and then the pH 5% probability by Turkey’s test, using SPSS program was adjusted to 6.8. Pancreatin was added and the sample (version 13). was incubated for 60 minutes at 40°C and the pH was ad- justed to 4.5. The solution was filtered and the insoluble Competing interests The authors declare that they have no competing interests. residue was washed with distilled water, 95% and 99% of ethanol and then dried overnight at 105°C, cooled and Authors’ contributions weighed (IDF). The filtrate was precipitated with hot 95% TM performed most of the experimental work, (apart from the mineral ethanol and filtered, washed with ethanol (78%, 95% and analysis), took part in the evaluation of the results and wrote the manuscript. All authors participated in the design of the study, supervised the field work 99%), dried, cooled and weighed (SDF). The results were and data collection and took part in the evaluation of the results. All authors corrected for protein and ash contents. read and approved the final manuscript.

Organic acids Acknowledgement This study was supported financially by the Swedish International Citric, malic, succinic and tartaric acids were analysed Development Agency Project “Technology Processing of Natural Resources using ion exchange chromatography (Metrohm Inter- for Sustainable Development” at Eduardo Mondlane University, Maputo, national, CH-9101, Herisau, Switzerland) with inverse Mozambique. We thank Dr Ulf Nilsson, Mr Christer Fahlgren and Mr Luqman Masood for their excellent technical assistance. suppression and conductivity detection. Reference sam- ples of the organic acids (Merck, Darmstadt, Germany) Author details μ 1Department of Biological Science, Science Faculty, Eduardo Mondlane were injected via a 20 l loop onto a column (250 mm × 2 μ University, PO Box 257, Maputo, Mozambique. Department of Chemistry, 7.8 mm, 9 m, polystyrene/divinylbenzene copolymer, Science Faculty, Eduardo Mondlane University, PO Box 257, Maputo, Metrosep 6.1005.200, Switzerland) and eluted isocra- Mozambique. 3Department of Food Technology, Engineering and Nutrition, tically at a flow rate of 0.5 mL min-1 and a pressure 3.8 Lund University, PO Box 124, Lund SE-221 00, Sweden. MPa. The suppressor system was regenerated by pum- Received: 15 October 2012 Accepted: 27 February 2013 ping a solution of 10 mM LiCl, together with Millipore Published: 8 March 2013 water, through the system. Different ratios of the mobile phase (0.5 mM sulphuric acid and acetone) were tested, References Al-Farsi M, Alasalvar C, Morris A, Baron M, Shahidi F (2005) Compositional and as well as different column temperatures (30, 40 and sensory characteristics of three native Sun-dried date (phoenix dactylifera L.) 50°C). The optimal conditions for separation of the or- varieties grown in Oman. J Agr Food Chem 53(19):7586–7591 ganic acids were 0.5 mM sulphuric acid and acetone Amarteifio JO, Mosase MO (2006) The chemical composition of selected indigenous fruits of Botswana. J Appl Sci Environ Manag 10(2):43–47 (85:15, v/v) and 30°C. Stock standard solutions (1 mg/mL) Anderson JM, Smith BM, Gustafson NJ (1994) Health benefits and practical of the organic acids were prepared and kept at +4°C. Dif- aspects of high-fiber diets. Am J Clin Nutr 59(suppl):1242S–1247S ferent dilutions of the stock solutions were used for the AOAC (2000) Horwitz W (ed) Official methods of analysis of AOAC International, 17th edn. Association of Analytical Communities, AOAC International, Md, USA calibration curves. Fruits collected in 2009 in Manhiça Arthey D, Ashurst PR (2001) Fruit Processing: nutrition, products and quality were used for this analysis. About 500 mg fruit pulps were management, 2nd edn. Aspen Publishers, USA mixed with 5 ml sulphuric acid (Ultra Turrax TP18/10) for Asp N-G, Johansson C-G, Hallmer H, Siljeström M (1983) Rapid enzymatic assay of insoluble and soluble dietary fiber. J Agric Food Chem 31:476–482 2 minutes and then centrifuged at 2000 rpm for 15 mi- Black R (2003) Micronutrient deficiency—an underlying cause of morbidity and nutes. The supernatants were filtered through 0.45 μm mortality. Bull World Health Organ 81(2):79, Published online March 25, 2003 membrane filters and diluted with Millipore water to ap- Bonjour JP, Gueguen L, Palacios C, Shearer MJ, Weaver CM (2009) Minerals and vitamins in bone health: the potential value of dietary enhancement. propriate concentrations. The retention time and peak Brit J Nutri 101(11):1581–1596 areas were compared with reference samples run under Brown L, Rosner B, Willett WW, Sacks FM (1999) Cholesterol-lowering effects of the same conditions and used to calculate the concen- dietary fiber: a meta-analysis1, 2. Am J Clin Nutri 69:30–42 Commission Decision (2008) Authorising the placing on the market of Baobab tration of the organic acids in the fruit samples. All ex- dried fruit pulp as a novel food ingredient under Regulation (EC) No 258/97 periments were performed at least in triplicates. of the European Parliament and of the Council. http://eur-lex.europa.eu/ LexUriServ/LexUriServ.do?uri=OJ:L:2008:183:0038:0039:EN:PDF Accessed at 05/ 06/2012 Minerals de Oliveira Sousa AG, Fernandes DC, Alves AM, de Freitas JB, Naves MMV (2011) Selected samples of the pulps, seeds and kernels were Nutritional quality and protein value of exotic almonds and nut from the prepared for determination of mineral content. The Brazilian Savanna compared to peanut. Food Research Inter 44(7):2319–2325 Effiong GS, Udo IF (2010) Nutritive Values of four Indigenous wild fruits in samples were digested in a microwave oven with con- Souteastern Nigeria. Elect J Env, Agric Food Chem 9(7):1168–1176 centrated nitric acid and then analysed by Inductively Ekesa BN, Walingo M, Abukutsa-Onyango MO (2009) Accessibility to and Coupled Plasma-Atomic Emission Spectrometry (ICP- composition of indigenous vegetables and fruits by rural households in mutungu division, western Kenya. Afr Food, Agric, Nutri Dev 9(8):1725–1738 AES, Perkin Elmer, OPTIMA 3000 DV). The experi- Elleuch M, Bedigian D, Roiseux O, Besbes S, Blecker C, Attia H (2011) Dietary ments were performed in duplicate. fibre and fibre-rich by-products of food processing: Characterisation, Magaia et al. SpringerPlus 2013, 2:88 Page 8 of 8 http://www.springerplus.com/content/2/1/88

technological functionality and commercial applications: A review. Food Reinhold JG, Garcia LJS, Garzon PMD (1981) Binding of iron by fiber of wheat Chem 124(2):411–421 and maize. Am J Clin Nutri 34:1384–1391 Eromosele IC, Eromosele O, Kuzhkuzha DM (1991) Evaluation of mineral elements Rodríguez R, Jiménez A, Fernández-Bolaños J, Guillén R, Heredia A (2006) Dietary and ascorbic acid contents in fruits of some wild plants. Plant Food Hum fibre from vegetable products as source of functional ingredients. Trends Nutri 41:151–154 Food Sci Technol 17(1):3–15 Falade OS, Olusoga RS, Oladipo A, Tubosun A, Adewusi SRA (2003) The level of Saka JDK, Msonthi JD (1994) Nutritional value of edible fruits of indigenous wild organic acids in some Nigerian fruits and their effect on mineral availability trees in Malawi. Forest Ecol Manag 64(2–3):245–248 in composite diets. Pak J Nutr 2(2):82–88 Sena LP, VanderJagt DJ, Rivera C, Tsin ATC, Muhamadu I, Mahamadou O, Millson FAO (2011) Nutrition Country Profile Republic of Mozambique. Food and M, Pastuszyn A, Glew RH (1998) Analysis of nutritional components of eight Agriculture Organization of the United Nations, Rome, 54 pp famine foods of the Republic of Niger. Plant Food Hum Nutri 52(1):17–30 Gillooly M, Bothwell TH, Torrance JD, MacPhail AP, Derman DP, Bezwoda WR, Van Duyn MAS, Pivonka E (2000) Overview of the health benefits of fruit and Mills W, Charlton RW, Mayet F (1983) The effects of organic acids, phytates vegetable consumption for the dietetics professional: Selected literature. and polyphenols on the absorption of iron from vegetables. Brit J Nutri J Am Dietetic Assoc 100(12):1511–1521 49:331–342 Glew RH, Ayaz FA, Sanz C, VanderJagt DJ, Huang HS, Chuang LT, Strnad M (2003) doi:10.1186/2193-1801-2-88 Changes in sugars, organic acids and amino acids in medlar (Mespilus Cite this article as: Magaia et al.: Dietary fiber, organic acids and germanica L.) during fruit development and maturation. Food Chem minerals in selected wild edible fruits of Mozambique. SpringerPlus 2013 83(3):363–369 2:88. Glew RH, VanderJagt DJ, Lockett C, Grivetti LE, Smith GC, Pastuszyn A, Millson M (1997) Amino acid, fatty acid, and mineral composition of 24 indigenous plants of Burkina Faso1. J Food Comp Anal 10:205–217 Glew RS, VanderJagt DJ, Huang YS, Chuang LT, Bosse R, Glew RH (2004) Nutritional analysis of the edible pit of Sclerocarya birrea in the Republic of Niger (daniya, Hausa). J Food Comp Anal 17(1):99–111 Hamilton CC, Geil PB, Anderson JW (1992) Management of obesity in diabetes mellitus. Diabetes Educ 18(5):407–410 Hasib A, Jaouad A, Mahrouz M, Khouili M (2002) HPLC determination of organic acids in Moroccan apricot. Cienc Tecnol Aliment 3(4):207–211 Ibrahim F (2011) Fruity response efficacy and fruit consumption among a group of civil servants of Oyo State, Nigeria. A J Clin Nutri 1(1):44–48 Ilich JZ, Brownbill RA, Tamborini L (2003) Bone and nutrition in elderly women: protein, energy, and calcium as main determinants of bone mineral density. Europ J Clin Nutr 57(4):554–565 Imelouane B, Tahri M, Elbastrioui M, Aouinti F, Elbachiri A (2011) Mineral contents of some medicinal and aromatic plants growing in eastern morocco. J Mater Environ Sci 2(2):104–111 Insel P, Ross D, McMahon K, Bernstein M (2011) Nutrition, Sudbury Massachusetts, 4th edn. Jones and Bartlett Publishers, USA Li BW, Andrews KW, Pehrsson PR (2002) Individual sugars, soluble, and insoluble dietary fiber contents of 70 high consumption foods. J Food Compos Anal 15(6):715–723 Lockett CT, Calvert CC, Grivetti LE (2000) Energy and micronutrient composition of dietary and medicinal wild plants consumed during drought. Study of rural Fulani, Northeastern Nigeria. Inter J Food Sci Nutri 51(3):195–208 López MAA, Martos FC (2004) Iron availability: An updated review. Inter J food Sci Nutri 55(8):597–606 Loredana L, Diehl H, Socaciu C (2006) HPLC Fingerprint of organic acids in fruit juices. Buletin USAMV-CN 62:288–292 Murray SS, Schoeninger MJ, Bunn HT, Pickering TR, Marlett JA (2001) Nutritional composition of some wild plant foods and honey used by hadza foragers of Tanzania. J Food Compos Anal 14(1):3–13 National Food Agency (2012) The National Food Administration’s food database. SLV, Sweden, http://www.slv.se/en-gb/ Accessed at 30/05/2012 National Research Council (2005) Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids (Macronutrients). The National Academies Press, Washington, http://www. nap.edu/books/0309085373/html/ Accessed 05/06/2012 New SA (2003) Intake of fruit and vegetables: implications for bone health. Submit your manuscript to a Proceed Nutr Soci 62:889–899 Nkafamiya II, Osemeahon SA, Dahiru D, Umaru HA (2007) Studies on the journal and benefi t from: chemical composition and physicochemical properties of the seeds of 7 Convenient online submission baobab (Adasonia digitata). African J Biotech 6(6):756–759 7 Nnam NM, Obiakor PN (2003) Effect of fermentation on the nutrient and Rigorous peer review antinutrient composition of baobab (adansonia digitata) seeds and rice 7 Immediate publication on acceptance (oryza sativa) grains. Ecol of Food and Nutri 42(4–5):265–277 7 Open access: articles freely available online Nour V, Trandafir I, Ionica ME (2010) HPLC organic acid analysis in different citrus 7 High visibility within the fi eld juices under reversed phase conditions. Not Bot Horti Agrobot Cluj-Napoca 7 38(1):44–48 Retaining the copyright to your article Osman MA (2004) Chemical and nutrient analysis of baobab (adansonia digitata) fruit and seed protein solubility. Plant Food Hum Nutri 59:29–33 Submit your next manuscript at 7 springeropen.com Paper III

Composition of amino acids, fatty acids and dietary fibre monomers in kernels of Adansonia digitata and Sclerocarya birrea

Telma Magaia1,2 and Kerstin Skog1

1 Department of Food Technology, Engineering and Nutrition, Lund University, PO Box 124, SE-221 00 Lund, Sweden 2 Department of Biological Science, Science Faculty, Eduardo Mondlane University, PO Box 257, Maputo, Mozambique Corresponding author: Telma Magaia, e-mail: [email protected]

Abstract

There is increasing worldwide demand for sources of energy and non-meat protein with balanced amino acid profiles. Nuts are generally rich in protein and essential amino acids, and have a high energy value due to their high fat content. Kernels from two wild fruits in Mozambique, Adansonia digitata and Sclerocarya birrea, were selected for this study based on their amino acid and fatty acid composition, as well as the monomeric composition of their dietary fibre. Information on these compounds in the kernels of these fruits is scarce in the literature. The fatty acid composition differed between the two kernels, but both are rich in unsaturated fatty acids. The dominating fatty acids in Adansonia digitata kernels were palmitic, linoleic and oleic acid; their contents varying from 25.7% to 34.9% of the total fatty acids. In Sclerocarya birrea kernels the main fatty acid was oleic acid (72.4%), found together with small amounts of linoleic, stearic and palmitic acid (6.8 to 12.1%). All common amino acids were detected in both types of kernels; glutamic acid being the most abundant amino acid, comprising more than 20% of the protein. The contents of essential amino acids in the kernels were compared with the amino acid requirements stated by the WHO for children aged 3-10 years. The findings indicate that A. digitata and S. birrea kernels can provide good, cheap sources of protein, especially when combined with foods with high lysine content. The results of this study show

1

that the intake of these kernels can help provide the fatty acids and amino acids required in the daily diet, especially for people living in rural areas. The results can be used for intake estimates, and to encourage increased consumption and utilization of these kernels.

Key words: Wild fruit, nutrients, consumption, Mozambique, seeds

Introduction

Wild fruits in Mozambique are an important source of food for rural people. Especially the seeds and kernels add essential nutrients to the diet, and are available when other foods are scarce [1]. The main staple foods in Mozambique are cassava, maize, beans, sorghum and rice. Cereals and starchy roots provide almost 80% of the dietary energy supply, while the contribution from pulses (mainly beans), nuts and oil crops is 5%. However, the diet in rural areas is poor in fat, protein and micronutrients, and does not supply enough energy to meet nutritional requirements [2]. Fat is an important component in the diet as it provides the body with energy in a concentrated form. In addition, dietary fats provide essential fatty acids and fat- soluble vitamins. Linoleic acid (an omega-6 fatty acid) and alpha-linolenic acid (an omega-3 fatty acid) are the essential fatty acids required in the diet [3]. Among the food sources of essential fatty acids are fish, soya oil, seed oils, kernels and nuts. Protein is another important component in the diet. Inadequate supply of protein is considered to be responsible for malnutrition among people living in developing countries. The content of essential amino acids determines the protein quality and the extent to which the protein can satisfy the amino acid requirements [4]. Kernels from wild fruits are commonly consumed in the southern and central parts of Mozambique, where a variety of edible fruits can be found. The kernels offer a convenient and cheap means of providing fat, protein, minerals and other health- promoting components to people living in rural areas [5]. We recently studied the content of fat, protein, dietary fibre and minerals in some wild fruits and kernels of Mozambique, and found that kernels of Adansonia digitata (baobab) and Sclerocarya birrea (marula) had high contents of fat, protein and dietary fibre [6]. Kernels of Adansonia digitata are highly appreciated in the diet, and are eaten roasted as a snack, or used for oil extraction. They are also used in soups or mixed with wild spinach or other food [7]. Kernels of Sclerocarya birrea can be dried and eaten alone, or cooked and, for example, eaten together with a mixture of dried peanut extracts, red pepper,

2 2

salt and other spices [8]. Literature data on the nutritional components of the kernels of Adansonia digitata and Sclerocarya birrea are scarce. The aims of the present study were, therefore, to interview a number of people to obtain information on how these kernels are used, and to measure the contents of amino acids, fatty acids and dietary fibre components in the kernels. It is important to have information on the nutritional composition of the kernels to be able to estimate adequate nutrient intake and to promote increased utilization and consumption of the kernels.

Materials and Methods

Samples

Kernels from two wild fruit species were studied: Adansonia digitata (A. digitata) (Fam. Bombacaceae, local name n´buyu or malambe), and Sclerocarya birrea (S. birrea) (Fam. Anacardiaceae, local name n´canhi). Seeds were collected in 2013 in districts where they are commonly consumed. Seeds (5 kg) from A. digitata were collected in family orchards in the Changara district, 95 km from the city of Tete. S. birrea seeds (1 kg), dried for two to three months, were obtained from a small family orchard in the Manhiça district, 50 km from Maputo. The seeds were crushed and the shells removed and the kernels inside were vacuum-packed in plastic bags and stored at - 18ºC in a freezer. Before analysis, A. digitata kernels were milled in a coffee grinder (TEFAL, Type 8100, Prep’Line, China) and sieved (500 µm mesh). S. birrea kernels were ground with a mortar and pestle. Chemicals

All chemicals were of analytical grade. Water was passed through a Milli-Q purification system. Amino acid standards, DL-norleucine (internal standard) and phenol, were purchased from Sigma Chemical Co. (St. Louis, USA), and lithium buffers and ninhydrin from Biochrom Ltd (Cambridge, UK). Sodium di-hydrogen phosphate, di-sodium hydrogen phosphate, sodium hydroxide, ethanol and acetone were purchased from Fluka (Sigma-Aldrich, Steinem, Germany). For the analysis of dietary fibre, pepsin was obtained from Merck (Darmstadt, Germany), pancreatin from Fluka (Sigma-Aldrich), and galacturonic acid, chloroform, monohydrate and myo-inositol (internal standard) from Sigma Chemical Co. The standards rhamnose, fucose, arabinose, xylose, mannose, galactose and glucose were obtained from Merck,

3

together with hydrochloric acid, sulphuric acid, acetic acid, acetic acid anhydride, ammonium hydroxide, potassium hydroxide, sodium sulphate, potassium borohydride and 1-methylimidazol.

4 4

Analysis

Fatty acids in the samples were analysed by an authorised laboratory using gas chromatography [9]. Determinations were performed in duplicate, and the difference between the determinations was generally below 3%. The results are expressed in g/100 g fat. About 5 g of each type of kernel was used for the analysis of amino acids. The samples were defatted using acetone in the proportions 1:10 (w/v), stirred for one hour and ultra-centrifuged at 18 000 rpm for 20 minutes. The samples were left to evaporate overnight at room temperature. Triplicate samples (35 mg) of kernels were put into glass tubes, 5 ml 6 N HCl containing 0.1% phenol was added, and the samples were hydrolysed at 110ºC for 24 hours [10, 11]. The samples were then centrifuged at 4000 rpm for 10 minutes, after which 2 ml of the supernatant was transferred to a round-bottomed flask and 100 µl of the internal standard (DL- norleucine 22.9 µmol/ml) was added. The samples were evaporated at 40ºC to dryness. Lithium citrate buffer (5 ml 0.2 M, pH 2.2) was then added, and the solutions were mixed for a few minutes, transferred to Eppendorf tubes with a 0.45 micron inner vial, and centrifuged at 13 000 rpm for 10 minutes. The samples were analysed using an amino acid analyser (Biochrom 30 series Amino Acid Analyser, Biochrom Ltd). The method is based on ion-exchange chromatography with post- column derivatization using ninhydrin. The injection volume was 20 µl. The column (Physiological Fluid High Performance, 200 x 4.6 mm, Biochrom Ltd) was eluted with lithium citrate buffers of pH 2.80, 3.00, 3.15, 3.50 and 3.55 at a flow rate of 25 ml/hour. Post-column derivatization was performed with ninhydrin at a flow rate 20 ml/hour. The temperature of the reaction coil was 135ºC. In the reaction coil, ninhydrin reacts with the amino acid present in the eluent and forms conjugated compounds with absorbance maxima at 570 and 440 nm. The internal standard was used to calibrate the amino acid response. Identification was confirmed by comparing the retention times of the peaks with those of a standard solution of amino acids run under the same conditions. All determinations were made in triplicate. The protein content in the defatted and untreated samples was determined using an elemental analyser (Flash EA 1112 Series, Thermo Fisher Scientific, Sweden) by means of combustion of 25 mg samples. Aspartic acid was used as a standard. The amount of protein was calculated by multiplying the amount of nitrogen by a factor 5.7. Determinations were performed in duplicate. The results of the amino acid analysis are reported as g amino acid/100 g protein. The dry matter content was determined after drying 2 g samples in an oven at 105 ºC until constant weight [9]. The determinations were performed in triplicate, and the

5

dry matter content was found to be 92.8 ± 0.0% for A. digitata and 95.8 ± 0.1% for S. birrea. The dietary fibre in the kernels was separated into soluble and insoluble fractions, and their monosaccharide compositions were analysed with gas chromatography (HP6890 Hewlett Packard, with a flame ionization detector) and the uronic acids with Visible spectrophotometry (Pharmacia Biotech Novaspec II model 80-2088-64, Sweden) using D-galacturonic acid monohydrate as a reference [12, 13]. Statistical analysis

Microsoft® Excel was used for the statistical evaluation. Student’s t-test was performed to determine significant differences. A value of p<0.05 was considered to indicate statistical significance. Interviews regarding traditional use of the kernels studied

When the seeds and kernels were collected in the different districts, local people were interviewed about common uses. The aim of the interviews was to obtain information on major consumers, how often the kernels were used for consumption, and examples of food preparation methods.

Results

The total fat content in the kernels was 31.7 ± 0.0% for A. digitata and 49.4 ± 3.7% for S. birrea. The results of the determination of the fatty acid compositions of the kernels are presented in Table 1, together with the fatty acid compositions of olive oil and peanuts for comparison. The unsaturated fatty acids constituted about 68 and 80% of the total fat content in A. digitata and S. birrea kernels, respectively. The most abundant fatty acids in A. digitata kernels were palmitic acid, linoleic acid and oleic acid; their contents varying from 25.7 to 34.9%. In S. birrea kernels, oleic acid constituted 72.4% of the fatty acids and linoleic, stearic and palmitic acid (6.8 to 12.1%). The difference in fatty acid composition between the kernels was significant for palmitic acid, oleic acid and linoleic acid (p<0.05).

6 6

A chromatogram from the analysis of amino acids in A. digitata is shown in Figure 1.

Minutes

Figure 1: Chromatogram from amino acid analysis of Adansonia digitata kernels. The peaks corresponding to the different amino acids are indicated.

The calibration curve was linear for the concentration range used in the analysis. Under acidic hydrolysis conditions, tryptophan is totally destroyed, while asparagine and glutamine are hydrolysed quantitatively to aspartic acid and glutamic acid, respectively. Cysteine is readily oxidized to cysteic acid, and methionine to methionine sulphoxide, which is partly lost during hydrolysis. Table 2 gives the amino acid compositions of the A. digitata and S. birrea kernels. All common amino acids except tryptophan were detected. Glutamic acid was the most abundant amino acid in both types of kernels, comprising more than 20% of the total amino acid content, followed by arginine. The differences between the contents of the various amino acids in the kernels were not significant. The protein content was 35.0 ± 0.0% in A. digitata and 29.2 ± 0.0% in S. birrea. After separation of the dietary fibre into soluble and insoluble fractions, the contents of monosaccharides and uronic acids were determined. The calibration curves for both components were linear over the concentration range used in the analysis. The results are given in Table 3. The compositions of the monomers (neutral

7

sugars) in the dietary fibre fractions differed between the kernels. The main constituent in the insoluble fraction of A. digitata kernels was glucose, followed by arabinose and uronic acids, while in the soluble fraction arabinose was the dominating component, followed by uronic acids. In kernels of S. birrea, uronic acids constituted more than 90% of both dietary fibre fractions. From the interviews, it was found that traditionally, the seeds of A. digitata are crushed, boiled and mixed with local plant food to make a sauce consumed with boiled maize flour. The seeds are also boiled with a small amount of water or milk to make porridge for children. The kernels of S. birrea were said to be delicious, adding flavour to the dishes. They are eaten as a snack, sometimes roasted, ground, mixed with water and local plant food and boiled to make a sauce. The kernels are used instead of peanuts in time of drought or in combination with peanuts. The kernels can also be crushed to produce oil for cooking.

Discussion

New nutritional data have been obtained for two commonly consumed kernels of wild fruits in Mozambique. Both types of kernels form part of the diet in rural areas and are highly appreciated, crushed, boiled and mixed with local plant foods. There are few reports on the nutritional value of these kernels, but our data and literature data show that the quality of both fat and protein is good. Differences between reported analytical data are probably due to the use of different analytical methods, and variability in the raw materials in terms of growth location, weather conditions, maturity, handling and storage. Our study shows that the fatty acid composition differed between the two kernels, in agreement with previous reports. For A. digitata kernels our results are in agreement with some recent reports [14, 15], but somewhat higher than in older reports [16, 17]. For S. birrea our results agree with literature data [18-21]. Both A. digitata oil and S. birrea oil are rich in monounsaturated fatty acids, and have been reported to be very stable [18, 20, 21]. A. digitata oil is very stable with a shelf life estimated to be between 2 and 5 years [22]. Interestingly, the fatty acid compositions of S. birrea kernels and olive oil are comparable; both having high contents of the monounsaturated fatty acid oleic acid, (>70%). Olive oil is generally regarded as having a good fat quality [3]. The consumption of monounsaturated fatty acids has been associated with decreased levels of low-density lipoprotein cholesterol and possibly increased high-density lipoprotein

8 8

cholesterol [23]. However, its ability to raise high-density lipoprotein cholesterol is still being debated. The fatty acid compositions of A. digitata kernels and peanuts correspond well, especially with regard to oleic acid and the essential linoleic acid. In addition, kernels of A. digitata contained low amounts, ~2%, of the essential linolenic acid. These essential fatty acids have many functions in the body, and are of importance, for example, in the immune system, cell membranes and for the function of the brain and skin [24]. They may also reduce the risk of heart disease [25]. Some authorities, e.g. the Food and Nutrition Board of the US Institute of Medicine, have established adequate intake (AI) levels, in grams, for omega-6 and omega-3 fatty acids [25], while the European Food Safety Authority (EFSA) and the World Health Organization (WHO) recommend an acceptable range of omega-3 and omega-6 fatty acid intake related to the energy intake [26, 27]. Table 4 gives the estimated contribution to the AI for omega-6 and omega-3 fatty acids from the consumption of 40 g kernels. It can be seen that 40 g A. digitata kernels can cover the daily intake of omega-6 fatty acids for those aged 4 to 13 years, and provide about 90% of the requirement for pregnant women. The same quantity of A. digitata kernels provides 60-93% of the AI of omega-3 fatty acids for the same groups. The contribution from 40 g S. birrea kernels is lower, and to reach the AI levels an intake of around 160 g kernels is needed. However, these kernels may not be the only source of essential fatty acids in the diet. Table 4 also gives the AI levels of protein and the contribution from the intake of 40 g kernels. For children aged 4 to 8 years, 74 and 61% of the AI is covered by A. digitata kernels and S. birrea kernels, respectively. For older children and pregnant women, the contribution is less, but it may be possible for these groups to increase their intake of kernels, especially if they are eaten as a snack. Furthermore, these two groups generally eat larger portions of dishes containing protein. The amino acid profiles given in Table 2 generally agree with reports on amino acids in whole seeds of A. digitata [14, 15] and kernels of S. birrea [28, 29]. However, our data for the S-amino acids are lower. This may be explained by partial breakdown of these amino acids during hydrolysis. Regarding the other essential amino acids, our results for lysine and valine in A. digitata are lower than in one previous report [14], and our findings regarding leucine, lysine and threonine in S. birrea are also lower [29]. It is interesting to note that the amino acid composition in the two kernels studied is comparable to that in peanuts and walnuts [30], which commonly form part of the diet. Ideally, the protein in the diet should provide the body’s requirement of all the essential amino acids, in the same relative proportions. The contents of essential amino acids in the kernels are compared with the amino acid requirement stated by

9

the WHO for children aged 3-10 years [31] in Table 5. Histidine is regarded as being essential for children, but not for adults. It should be noted that, according to the WHO, the sum of phenylalanine and tyrosine is important, although tyrosine is a non-essential amino acid. The amounts of methionine + cysteine were lower than the requirement in both kernels, but it is reasonable to believe that the real content is higher than the analytical value, due to losses during sample preparation, as described above. The amount of lysine is lower than that required in both A. digitata and S. birrea kernels, while in S. birrea kernels leucine is also lower than required. However, the total amino acid composition indicates that A. digitata and S. birrea kernels can provide good, cheap sources of protein, especially if combined with foods with higher contents of lysine. In conclusion, our findings indicate that kernels of A. digitata and S. birrea form part of the diet in rural areas in Mozambique, and that they are good sources of protein and essential amino acids, as well as fat and essential fatty acids. They are rich in unsaturated fatty acid, and A. digitata kernels contain appreciable amounts of essential fatty acids, while the content in S. birrea is lower. We also conclude that the kernels of A. digitata and S. birrea have a potential as sources of essential nutrients and can contribute to the nutritional needs of various communities. Furthermore, the results can be used for the estimation of dietary intake and to encourage increased consumption and use of these wild fruit kernels.

Acknowledgements

This study was supported financially by the Swedish International Development Agency’s Project “Technology Processing of Natural Resources for Sustainable Development”, at Eduardo Mondlane University, Maputo, Mozambique. We would like to thank Mr Christer Fahlgren, Dr Lina Haskå and Mrs Bindu Ayoor for their excellent technical assistance.

10 10

References

1. Van Wyk B, Van Wyk P, Van Wyk BE Photo Guide to Trees of Southern Africa, 2nd. Gauteng: Briza; 2008. 2. FAO Mozambique Nutrition Profile – Nutrition and Consumer Protection Division. In.; 2011.

3. National Food Agency The National Food Administration food database. Sweden http://www.slv.se/en-gb/ Accessed 22/11/2014.

4. FAO/WHO/UNU Energy and Protein Requirements: Past work and Future Prospects at the International Level In. Rome: WHO; 1981.

5. Eromosele I, Eromosele C, Kuzhkuzha D Evaluation of mineral elements and ascorbic acid contents in fruits of some wild plants. Plant Foods Hum Nutr 1991; 41(2):151-154.

6. Magaia T, Uamusse A, Sjöholm I, Skog K Proximate Analysis of Five Wild Fruits of Mozambique. Scientific World Journal 2013; http://dx.doi.org/10.1155/2013/601435:6 pages.

7. Glew R, VanderJagt D, Lockett C, Grivetti L, Smith G, Pastuszyn A, Millson M Amino Acid, Fatty Acid, and Mineral Composition of 24 Indigenous Plants of Burkina Faso. J Food Compost Anal 1997; 10(3):205-217.

8. den Adel S Use of marula products for domestic and commercial purposes by households in North- Central Namibia. In. Edited by 2.2 A. Namibia; 2002.

9. Horwitz W Official methods of analysis of AOAC International. 17th edn; 2000.

10. Stenberg M, Marko-Varga G, Öste R Enantioseparation of d- and l-amino acids by a coupled system consisting of an ion-exchange column and a chiral column and determination of d-aspartic acid and d-glutamic acid in soy products. Food Chem 2002; 79(4):507-512.

11. Fermin BC, Radinsky JA, Kratochvil RJ, Hall JE, Lo YM Integration of Rapid Derivatization and Gradient Elution Techniques for Enhanced High-Performance Liquid Chromatography Analysis of Key Amino Acids in Wheat Flour. J Food Sci 2003; 68(9):2667-2671. 12. Theander O, Aman P, Westerlund E, Andersson R, Pettersson D Total dietary fiber determined as neutral sugar residues, uronic acid residues, and Klason lignin (the Uppsala method): collaborative study. J AOAC Int 1995; 78(4):1030-1044.

11

13. Haskå L, Nyman M, Andersson R Distribution and characterisation of fructan in wheat milling fractions. J Cereal Sci 2008; 48(3):768-774. 14. Osman M Chemical and Nutrient Analysis of Baobab (Adansonia digitata) Fruit and Seed Protein Solubility. Plant Foods Hum Nutr 2004; 59(1):29-33. 15. Ezeagu IE BAOBAB (Adansonia digitata L.) Seed protein utilization in young albino rats I: Biochemical ingredients and performance characteristics. Anim Res Int 2005; 2(1):240-245. 16. Eteshola E, Oraedu A Fatty Acid Compositions of Tigernut Tubers (Cyperus esculentus L.), Baobab Seeds (Adansonia digitata L.), and Their Mixture. JAOCS 1996; 73:255-257. 17. Ezeagu I, Petzke K, Lange E, Metges C Fat Content and Fatty Acid Composition of Oils Extracted from Selected Wild-Gathered Tropical Plant Seeds from Nigeria. JAOCS 1998; 75:1031-1035. 18. Mariod A, Matthäus B, Eichner K Fatty Acid, Tocopherol and sterol Composition as well as Oxidative Stability of three Unusual Sudanese Oils J Food Lipid 2004; 11(3):179-189. 19. Zharare P, Dhlamini N Characterization of Marula (Sclerocarya caffra) Kernel Oil and Assessment of its potential use in Zimbabwe. J Food Techn Africa 2000; 5(4):126-128. 20. Zimba N, Wren S, Stucki A Three major tree nut oils of southern central Africa: Their uses and future as commercial base oils. The International Journal of Aromatherapy 2005; 15:177-182. 21. Kleiman R, Ashley D, Brown J Comparison of two seed oils used in cosmetics, moringa and marula. Ind Crops Prod 2008; 28(3):361-364. 22. Vermaak I, Kamatou GPP, Komane-Mofokeng B, Viljoen AM, Beckett K African seed oils of commercial importance Cosmetic applications. South African Journal of Botany 2011; 77(4):920-933. 23. Coultate TP Food: The Chemistry of its Components, 5th edn. Cambridge UK: Royal Society of Chemistry; 2009. 24. Uauy R, Hoffman D, Peirano P, Birch D, Birch E Essential fatty acids in visual and brain development. Lipid 2001; 36(9):885-895. 25. Food and Nutrition Board Dietary Fats: Total Fat and Fatty Acids. Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids. In.: http://www.nap.edu/books/0309085373/html/ 2002: 422-541. 26. EFSA Scientific Opinion on Dietary Reference Values for fats, including saturated fatty acids, polyunsaturated fatty acids, monounsaturated fatty acids, trans fatty acids, and cholesterol. vol. 8; 2010: 1459. 27. Joint WHO/FAO Interim Summary of Conclusions and Dietary Recommendations on Total Fat & Fatty Acids. In: Joint FAO/WHO Expert Consultation on Fats and Fatty Acids in Human Nutrition Geneva: FAO; 2008: 1-14.

12 12

28. Glew RS, VanderJagt DJ, Huang YS, Chuang LT, Bosse R, Glew RH Nutritional analysis of the edible pit of Sclerocarya birrea in the Republic of Niger (daniya, Hausa). J Food Compost Anal 2004; 17(1):99-111. 29. Muhammad S, Hassan LG, Dangoggo SM, Hassan SW, Umar KJ, Aliyu RU Nutritional and antinutritional composition of Sclerocarya birrea seed kernel Studia Universitatis “Vasile Goldiş”, Seria Ştiinţele Vieţii 2011; 21, (4):693-699. 30. National Nutrient Database for Standard Reference Release 27. 31. WHO/FAO/UNU Protein and amino acid requirements in human nutrition In: Report of a Joint WHO/FAO/UNU Expert Consultation, Agriculture Organization of the United Nations 935. 2007.

13

Table 1. Fatty acid composition in kernels of A. digitata and S. birrea (mean ± standard deviation, n=2). The contents in olive oil and peanuts are given for comparison Fatty acid A. digitata S. birrea Olive oil1 Peanuts1 No. of carbon Systematic atoms and double name bonds (g/100 g fat) C14:0 Myristic 0.2 ± 0.0 n.d.2 - 0.4 C16:0 Palmitic 25.7 ± 0.2 12.1 ± 0.4 11.1 10.4 C16:1 Palmitoleic 0.2 ± 0.0 0.2 ± 0.0 1.2 0.0 C17:0 Margaric 0.2 ± 0.0 0.1 ± 0.0 - - C17:1 Margaroleic 0.2 ± 0.0 n.d. - - C18:0 Stearic 4.6 ± 0.1 7.3 ± 0.1 2.7 2.2 C18:1 Oleic 34.9 ± 0.7 72.4 ± 0.6 70.7 48.3 C18:2 Linoleic3 29.9 ± 0.9 6.8 ± 0.4 8.1 31.7 C18:3 Linolenic3 2.1 ± 0.0 n.d. 0.7 0.0 C20:0 Arachidic 0.9 ± 0.1 0.6 ± 0.0 - 0.0 C20:1 Eicosenoic 0.2 ± 0.0 0.3 ± 0.0 - - C22:0 Behenic 0.3 ± 0.0 0.1 ± 0.0 - - C22:1 Erucic 0.1 ± 0.0 n.d. - - C24:0 Lignoceric 0.2 ± 0.0 0.2 ± 0.0 - - 1 Data recalculated from The Swedish National Food Administration’s food database, version 28/02/2014. 2 n.d. = not detected 3Essential fatty acid

14 14

Table 2. Amino acid composition in kernels of A. digitata and S. birrea (mean ± standard deviation, n=3). The contents in peanuts and walnuts are given for comparison Amino acid A. digitata S. birrea Peanuts1 Walnuts1 (g/100 g protein) Cysteine* 0.5 ± 0.0 0.7 ± 0.0 1.3 2.4 Histidine* 2.2 ± 0.1 2.5 ± 0.2 2.5 2.5 Isoleucine* 3.4 ± 0.2 3.9 ± 0.3 3.5 3.9 Leucine* 6.0 ± 0.4 4.3 ± 1.3 6.5 6.4 Lysine* 3.4 ± 1.0 2.9 ± 0.2 3.6 2.7 Methionine* 0.8 ± 0.1 0.7 ± 0.1 1.2 1.9 Phenylalanine* 4.5 ± 0.3 4.6 ± 0.3 5.2 4.4 Threonine* 2.8 ± 0.1 2.8 ± 0.2 3.4 3.4 Valine* 4.5 ± 0.3 5.0 ± 0.4 4.2 5.0 Alanine 3.8 ± 0.3 3.7 ± 0.3 3.9 4.2 Arginine 11.4 ± 0.6 13.9 ± 1.1 12.0 14.6 Aspartic acid 8.1 ± 0.4 8.4 ± 0.5 12.2 10.2 Glycine 4.4 ± 0.3 4.9 ± 0.3 6.0 5.2 Glutamic acid 23.7 ± 1.3 25.7 ± 1.8 20.8 19.5 Proline 3.2 ± 0.2 3.5 ± 0.2 4.4 3.8 Serine 4.6 ± 0.3 2.9 ± 0.3 4.9 5.4 Tyrosine 2.5 ± 0.1 4.7 ± 0.2 4.1 3.0 Total 89.8 95.9 99.7 98.5 1 Data recalculated from USDA, Nutrient Database for Standard Reference, release 27, accessed on 22/11/ 2014. * Essential amino acid

15

Table 3. Monomeric composition of dietary fibre in kernels of A. digitata and S. birrea (mean ± standard deviation, n=2) A. digitata S. birrea (%) Insoluble Soluble Insoluble Soluble dietary fibre dietary fibre dietary fibre dietary fibre Rhamnose 1.9 2.6 n.d. n.d. Fucose n.d. n.d. n.d. n.d. Arabinose 25.9 37.6 3.9 1.3 Xylose 11.1 5.5 3.2 1.3 Mannose 3.0 0.4 n.d. n.d. Galactose 5.6 9.8 n.d. 1.2 Glucose 37.6 11.1 10.5 n.d. Uronic acids 14.8 33.0 82.4 96.2 n.d. = not detected

16 16

Table 4. Adequate intakes (AI) of essential fatty acids and protein for different life stage groups and contribution to the AI from the consumption of 40 g A. digitata or S. birrea kernels A. digitata S. birrea Nutrient Life stage group AI (g/day)1 Contribution (% of AI) Omega-6 fatty acids Children 4 to 8 years 10 120 27 Males 9 to 13 years 12 100 23 Females 9 to 13 years 10 120 27 Pregnant women 13 92 21 Omega-3 fatty acids Children 4 to 8 years 0.9 93 -

Males 9 to 13 years 1.2 70 - Females 9 to 13 years 1.0 84 - Pregnant women 1.4 60 - 19 74 61 Protein Children 4 to 8 years 9 to 13 years 34 41 34 Pregnant women 71 20 16 1 Data from Food and Nutrition Board, 2002 [25]

17

Table 5. Essential amino acids in kernels of Adansonia digitata and Sclerocarya birrea compared with the WHO protein requirement for children aged 3-10 years old.

Amino acid A. digitata S. birrea WHO1 (g/100 g protein)

Histidine 2.2 2.5 1.6

Isoleucine 3.4 3.9 3.0

Leucine 6.0 4.3 6.1

Lysine 3.4 2.9 4.8

Methionine + cysteine 1.3 1.4 2.3

Phenylalanine + tyrosine 7.0 9.3 4.1

Threonine 2.8 2.8 2.5

Valine 4.5 4.9 4.0 1 Data from WHO, 2007 [31]

18 18

Paper IV

Effect of heating, lactobacillus fermentation and enzyme treatment on the content of phytic acid in kernels of Adansonia digitata and Sclerocarya birrea

Telma Magaia1,2, Ingegerd Sjöholm1, Estera Dey3 and Kerstin Skog1

1 Department of Food Technology, Engineering and Nutrition, Lund University, PO Box 124, SE-221 00 Lund, Sweden 2 Department of Biological Science, Science Faculty, Eduardo Mondlane University, PO Box 257, Maputo, Mozambique 3 Division of Pure and Applied Biochemistry, Lund University, PO Box 124 SE-221 00 Lund, Sweden

Corresponding author: Telma Magaia e-mail: [email protected] Abstract

Adansonia digitata and Sclerocarya birrea are wild trees found in many African countries, the leaves, fruits and kernels of which are commonly eaten in rural areas of Mozambique. The kernels contain high amounts of several minerals, for example iron, magnesium and zinc. However, these kernels have been shown to contain phytic acid, which may decrease the absorption of minerals from the diet, especially zinc and iron, and to a lesser extent, also calcium and magnesium. In this study we have analysed the content of phytic acid in kernels from Adansonia digitata and Sclerocarya birrea using high-performance ion chromatography. Our results give the first data on contents of phytic acid in these kernels from Mozambique. The results show that the amount of phytic acid in the kernels can be reduced by various processing techniques such as boiling or autoclaving followed by incubation with Lactobacillus plantarum or by enzyme (phytase) treatment. The content of phytic acid was lower after autoclaving than after boiling. Treatment with phytase reduced the phytic acid

1

content by 20-30% after only 15 minutes. Around half the content of Fe, Mg and Zn in the kernels was found in the supernatant after 15 minutes’ enzyme treatment.

Key words: Wild fruits, Mozambique, phytase, high-performance ion chromatography 1.0 Introduction

Kernels from wild fruits are commonly eaten in rural areas of Mozambique. They provide essential nutrients to the diet and are available in the dry season when other foods are scarce (Jama et al., 2008; Mangue and Oreste, 1999). The kernels from the Adansonia digitata (A. digitata) fruit are commonly eaten fresh or roasted, or they can be dried and ground into flour which can be added to soups and stews as a thickener, roasted and ground into a paste, or boiled for a long time, fermented and then dried for use (De Caluwé et al., 2009; Nnam and Obiakor, 2003). Kernel sauce is prepared by (optional) roasting of the kernels followed by grinding, and the resulting product is used as protein concentrates in tomato sauces or other spiced sauces (Chadare et al., 2009). The kernels from the Sclerocarya birrea fruit (S. birrea) are very tasty and widely eaten. They can be dried and eaten alone, or cooked and eaten together with a mixture of dried peanut extract, red pepper, salt and other spices in the form of a “meat bundle” (Glew et al., 1997). In a previous study (Magaia et al., 2013), we found that kernels of A. digitata and S. birrea contained high amounts of several minerals, for example, iron, magnesium and zinc, as has also been reported in other studies (Emmanuel et al., 2011; Glew et al., 2004; Oyeleke et al., 2012; Wairagu et al., 2013). However, these kernels have been shown to contain phytic acid (for literature references, see Table 4), which may decrease the absorption of minerals from the diet (Holzapfel and Schillinger, 2002).

Phytic acid, also known as inositol hexaphosphate, has six PO4 groups and is a strong chelator to divalent ions, especially zinc and iron, and to a lesser extent, also calcium and magnesium (Greiner and Konietzny, 1998). Phytic acid is the principal storage form of phosphorus in many plant tissues, especially bran and seeds (Anastasio et al., 2010). Removal of phosphate groups from the inositol ring decreases the mineral binding strength of phytic acid and thus improves the nutritional value (Kumar et al., 2010). Domestic food preparation techniques, such as boiling, can reduce the phytic acid content to some extent, but soaking in an acid medium, lactic acid fermentation, sprouting, and the use of enzymes (phytase) are more effective methods (Greiner and

2 2

Konietzny, 2006; Liang et al., 2008; Mahgoub and Elhag, 1998; Sharma and Sehgal, 1992). Fermentation, either spontaneous or with a starter culture, is a simple and efficient means of reducing the phytic acid content in foods. At suitable pH (4.8 to 5.6) native plant phytase is activated, which removes phosphate groups from the phytic acid (Reale et al., 2007). Microbial phytase can be produced by Lactobacillus bacteria and the optimal pH is 5 to 6 (Greiner and Konietzny, 2006; Sandberg and Svanberg, 1991). Lactobacillus strains have been reported to produce extracellular phytase (Sreeramulu et al., 1996). Results from preliminary experiments (data not shown) indicated that the content of phytic acid was about 5.5% in kernels of A. digitata and 3.4% in S. birrea kernels. This prompted us to study how the content of phytic acid in kernels of A. digitata and S. birrea could be influenced by heating, fermentation with Lactobacillus bacteria or incubation with phytase. Furthermore, we determined the content of some minerals in the water solution after the enzyme treatment. 2.0 Materials and Methods

2.1 Samples

Kernels from two wild fruit species were studied: A. digitata (Fam. Bombacaceae, local name n´buvu or malambe), and S. birrea (Fam. Anacardiaceae, local name n´canhi). The kernels, collected in 2013, were sub-samples from another study (Magaia et al, Unpublished results). The hard shell of the A. digitata fruit was removed by crushing and the kernels were collected, vacuum-packed in different batches and stored at - 18°C in a freezer. Before analysis, A. digitata kernels were milled in a coffee grinder (TEFAL, Type 8100, PreP’Line, China) and sieved (500 µm mesh), and S. birrea kernels were ground with a mortar and pestle. 2.2 Chemicals

All chemicals were of analytical grade and de-ionised water was used for all experiments. Hydrochloric acid and Fe(NO3)3·9H2O were obtained from Fluka (Sigma-Aldrich, Steinheim, Germany) and sodium phytate from BDH Chemicals Ltd (UK). Wheat phytase (0.01-0.004 units/mg) was obtained from Sigma Chemical Co. (Stockholm, Sweden). For the experiments with Lactobacillus bacteria, the chemicals were obtained from Merck (Darmstadt, Germany), Sigma-Aldrich Co. (Steinheim,

3

Germany), and VWR-International (Stockholm, Sweden), and Lactobacillus plantarum 299v (DSM 9843, Probi AB, Lund, Sweden) from the local pharmacy. 2.3 Processing

The samples were subjected to boiling or autoclaving followed by fermentation with Lactobacillus bacteria or incubation with phytase without pre-treatment. In the first series of experiments, one capsule of Lactobacillus plantarum 299v was suspended in sterile MRS medium and incubated for 24 hours at 37°C. The cells were washed 3 times with small portions of 0.9% (w/v) NaCl and centrifuged at 4000 rpm for 10 min. L. plantarum cells were suspended in 50 ml 0.9% (w/v) NaCl and subjected to serial dilution. From each dilution tube 0.1 ml was plated onto MRS plate count agar. After incubation for 48 hours at 37°C, the number of colonies was counted to evaluate the cell level in the undiluted suspension; the result being 12 x 108 colony-forming units (CFU) per ml. To test the influence of L. plantarum on the phytic acid level in the kernels, 1 g sample was dissolved in 5 ml distilled water, and the samples were first either boiled or autoclaved for 15 minutes to inactivate possible intrinsic phytase. When the samples had cooled down, they were inoculated with 1 ml bacteria solution (12 x 108 CFU/ml) and fermented for 48 hours at 37°C. After fermentation the pH had decreased slightly to between 4 and 5. The samples were freeze-dried for later phytic acid analysis. In the second series of experiments, a stock solution of phytase (10 mg/ml) was prepared. Samples of 1 g kernel were suspended in 9 ml water and the pH adjusted to 5 (optimal pH for phytase). Then 1 ml of the stock solution was added and the samples were incubated in a water bath at 55°C for 2, 15, 30, 60 and 240 minutes. In addition, one sample was dissolved in 5 ml water and 5 ml of the stock solution. After enzymatic incubation, the samples were immediately boiled for 5 minutes. As a control, a sample without phytase was boiled for 5 minutes. The samples were cooled and half of the samples were freeze-dried and stored at -18ºC until phytic acid analysis. The other samples were centrifuged at 18000 rpm for 20 minutes, and selected supernatants were sent to the department of Ecology at Lund University, Sweden for determination of the mineral content. 2.4 Analysis

The dry matter content was determined after drying 2 g samples in an oven at 105°C until constant weight (AOAC, 2000). The determination was performed in triplicate.

4 4

Phytic acid was analysed as inositol hexaphosphate (Carlsson et al., 2001). Duplicate samples (0.5 g) were extracted with 20 ml 0.5 M HCl for 3 hours at 20°C under magnetic stirring. The extracts were frozen overnight, thawed and centrifuged at 12000 rpm for 10 minutes, and an aliquot (300 µl) of the supernatant was injected into a high-performance ion chromatography (HPIC) instrument (Waters Associates Inc, Milford, MA) equipped with a guard-column (PA-100, 4 x 50 mm i.d., Dionex Corp., Sunnyvale, CA, USA) and an analytical column (HPIC OmmiPac PA-100, 4 x 250 mm i.d.). The column was eluted isocratically with 80% HCl (1 M) and 20% water. Inositol hexaphosphate was identified and quantified after post-column reaction with Fe(NO3)3·9H2O, and the absorbance was monitored at 290 nm (Waters 486, tunable absorbance detector). Sodium phytate was used as an external standard. The determinations were performed in duplicate. The mineral content in the supernatant was determined after the addition of 65% nitric acid, corresponding to 1% of the sample volume. The samples were filtered and an aliquot was analysed using inductively coupled plasma-atomic emission spectrometry (ICP-AES, Perkin Elmer, OPTIMA 3000 DV). The experiments were performed in duplicate. 2.5 Statistical analysis

Microsoft® Excel was used for statistical evaluation. Student’s t-test was performed to determine significant differences. A value of p<0.05 was considered to indicate statistical significance. 3.0 Results

A chromatogram from the analysis of phytic acid in A. digitata kernels after phytase treatment is shown in Figure 1. The retention time for the peak corresponding to the inositol hexaphosphate was around 5 minutes. The calibration curve was linear for the concentration range used in the analysis. The results of the determinations of phytic acid in the first series of experiments with boiling and autoclaving followed by Lactobacillus fermentation are given in Table 1, where it can be seen that the content of phytic acid is lower after autoclaving than after boiling. The reduction in the amount of phytic acid after fermentation was significant for the autoclaved samples, but not for the boiled samples. Table 2 gives the amount of phytic acid in the kernels after treatment with phytase for different times. For A. digitata kernels, the phytic acid content after 2 minutes’

5

incubation was only somewhat lower than in the untreated kernels, but a slight decrease was observed after 15 minutes. This value was maintained up to 60 minutes. However, after 4 hours’ incubation, the content had decreased to around 63% of the original value, and this decrease was significant. Similar results were obtained for S. birrea kernels, and after 4 hours’ incubation the phytic acid content was only about 42% of the original value; this decrease was significant. The lowest phytic acid contents were seen after 60 minutes’ incubation with a 5-fold higher enzyme concentration, when the value decreased significantly to 48% and 33% of the initial value in A. digitata and S. birrea kernels, respectively. The results of the analysis of the supernatant of some of the incubated samples regarding their contents of iron, magnesium and zinc are given in Table 3. The results show that the amount of minerals in the supernatant was higher than in the samples not incubated, apart from zinc for the A. digitata sample. The value for zinc was unexpectedly high and could not be explained. The time for enzyme incubation did not seem to have a major influence on the mineral content, although there was a tendency for the longest incubation time to result in an increase in the concentration of magnesium in the supernatants from both kernels. 4.0 Discussion

This study is part of an on-going investigation of some wild fruits and kernels in Mozambique and the contents of compounds that are important from a nutritional point of view. Few data are available in the literature on phytic acid in the kernels studied here, and are summarized in Table 4. We have not found any data of phytic acid in these kernels from Mozambique. This means that our data will make a substantial contribution to the knowledge concerning phytic acid content in kernels from these two wild fruits. It is difficult to compare literature data on phytic acid as the results are sometimes expressed in terms of fresh weight, and sometimes based on dry matter content, while in other cases no information is given about fresh or dry weight. Furthermore, different analytical methods may have been used, which are sometimes not clearly described. The differences may also be explained to some extent by harvest year, location, climate and weather conditions. Based on the analyses of phytic acid conducted using HPIC, our results for heated A. digitata kernels (3.6 – 4.4 g/100 g) generally agree with the results reported from other studies using other methods (Table 4). Our results are, however, lower than results from previous studies (Adubiaro et al., 2011; Mitchikpe et al., 2008) and higher than in others (Ezeagu, 2005; Nkafamiya et al., 2007; Proll et al., 1998;

6 6

Saulawa et al., 2014). In addition, there is one report showing markedly lower values of phytic acid (Osman, 2004). Our results regarding the amounts of phytic acid in heated S. birrea kernels (1.8 – 2.2 g/100 g) are around five times higher than that found in another study (Muhammad et al., 2011). It is interesting to note that the phytic acid content in peanuts, which are commonly consumed, has been reported to be around 2 g/100 g (Lazarte et al., 2015). The addition of L. plantarum bacteria had a significant effect on the phytic acid content in the kernels which had been autoclaved before incubation. A slight reduction of the of phytic acid content in A. digitata seeds was reported after 72 hours fermentation (Nnam and Obiakor 2003) by the microflora on the seeds. Lactobacillus strains have been reported to produce extracellular phytase (Sreeramulu et al., 1996), which can reduce the phytic acid content (Reale et al., 2007). During soaking or enzyme incubation, phytic acid is transferred to the surrounding liquid. The addition of phytase to the medium reduced the phytic acid content in the supernatants by 40-60% after 4 hours’ treatment, but a five-fold higher enzyme concentration was more efficient. Thus future experiments should include different ratios of solids to liquid and different enzyme concentrations. Results from a previous study showed that the phytic acid content in A. digitata kernel flour decreased by 65.57% after boiling for 1 hour, and changing the water at 20 minutes’ intervals (Saulawa et al., 2014). This also suggests that the ratio of solid to liquid is an important factor for the reduction of phytic acid in the kernels. Furthermore, this indicates that boiling for prolonged period may decrease the phytic acid content. Changing the water may reduce the content of important minerals, as we observed that the amounts of iron, magnesium and zinc in the supernatant increased after enzyme incubation, and this would probably also be the case in boiling. As can be seen from Table 3, almost 50% of the original mineral content can be found in the supernatant after only a few minutes’ enzyme incubation, apart from zinc in A. digitata (20%). In this study, we observed a weak relation between the content of phytic acid and minerals in the supernatant of the enzyme-treated samples, although the incubation time had no significant effect on the mineral content. One explanation of this could be that part of the enzyme was bound to the kernel matrix, reducing the effect. 4.1 Conclusions

New data on the content of phytic acid in kernels from two wild fruits have been obtained. The results show that the presence of phytic acid in the kernels can be reduced by various processing techniques. The addition of bacteria had significant

7

effect on the phytic acid content in the autoclaved kernels. Enzyme treatment reduced the phytic acid content by 20-30% after 15 minutes, and a higher enzyme concentration was found to be more efficient. Almost 50% of the estimated original content of minerals was found in the supernatant after a few minutes’ enzyme incubation. This suggests that the water used for boiling should be included in the dishes if the kernels are to be used as a dietary supplement. Acknowledgements

This study was financially supported by the Swedish International Development Agency Project “Technology Processing of Natural Resources for Sustainable Development” at Eduardo Mondlane University, Maputo, Mozambique. We would like to thank Mrs Serafina Vilanculos and Mrs Annette Almgren at the Department of Food Science, Chalmers University of Technology, Gothenburg, Sweden for excellent technical assistance.

Figure legend

Figure 1: Chromatogram from the analysis of A. digitata kernels after phytase treatment. The retention time for the peak corresponding to inositol hexaphosphate is around 5 minutes. References

Adubiaro, H. O., Olaofe, O., Akintayo, E. T., and Babalola, O. O. (2011). Chemical composition, calcium, zinc and phytate interrelationships in baobab (Adansonia digitata) seed flour. Advance Journal of Food Science and Technology 3 (4), 228-232. Anastasio, M., Pepe, O., Cirillo, T., Palomba, S., Blaiotta, G., and Villani, F. (2010). Selection and use of phytate-degrading LAB to improve cereal-based products by mineral solubilization during dough fermentation. Journal of Food Science 75 (1), M28-M35. AOAC (2000). Officiall methods of analysis of AOAC International, (17 th ed.) Md; USA. Chadare, F. J., Linnemann, A. R., Hounhouigan, J. D., Nout, M. J. R., and Van Boekel, M. A. J. S. (2009). Baobab Food Products: A review on their composition and nutritional value. Critical Reviews in Food Science and Nutrition 49, 254-274. Carlsson, N.-G., Bergman, E.-L., Skoglund, E., Hasselblad, K., and Sandberg, A.-S. (2001). Rapid analysis of inositol phosphates. Journal of Agricultural and Food Chemistry 49, 1695-1701.

8 8

De Caluwé, E., De Smedt, S., Assogbadjo, A. E., Samson, R., Sinsin, B., and Van Damme, P. (2009). Ethnic differences in use value and use patterns of baobab (Adansonia digitata L.) in northern Benin. African Journal of Ecology 47, 433-440. Emmanuel, T. V., Njoka, J. T., Catherine, L. W., and Lyaruu, H. M. (2011). Nutritive and anti-nutritive qualities of mostly prefereed edible woody plants in selected drylands of Iring District, Tanzania. Pakistan Journal of Nutrition 10 (8), 786-791. Ezeagu, I. E. (2005). Baobab (Adansonia digitata L.) Seed protein utilization in young albino rats. Biochemical ingredients and performance characteristics. Animal Research International 2 (1), 240-245. Glew, R.H., VanderJagt, D.J., Lockett, C., Grivetti, LE., Smith, G.C., Pastuszyn, A., and Millson, M. (1997). Amino acid, fatty acid, and mineral composition of 24 indigenous plants of Burkina Faso. Journal of Food Composition and Analysis 10, 205-217. Glew, R. S., VanderJagt, D. J., Huang, Y. -S., Chuang, L. -T., Bosse, R., and Glew, R. H. (2004). Nutritional analysis of the edible pit of Sclerocarya birrea in the Republic of Niger (daniya, Hausa). Journal of Food Composition and Analysis 17, 99-111. Greiner, R., and Konietzny, U. (1998). Endogenous phytate-degrading enzymes are responsible for phytate reduction while preparing beans (Phaseolus vulgaris). Journal of Food Processing Preservation 29, 321-331. Greiner, R., and Konietzny, U. (2006). Phytase for food application. Food Technology and Biotechnology 44 (2), 125-140. Holzapfel, W. H., and Schillinger, U. (2002). Introduction to pre-and probiotics. Food Research International 35, 109-116. Jama, B. A., Mohamed, A. M., Mulatya, J., and Njui, A. N. (2008). Comparing the “Big Five”: A framework for the sustainable management of indigenous fruit trees in the drylands of East and Central Africa. Ecological Indicators 8, 170-179. Kumar, V., Sinha, A. K., Makkar, H. P. S., and Becker, K. (2010). Dietary roles of phytate and phytase in human nutrition: A review. Food Chemistry 120, 945-959. Lazarte, C. E., Carlsson, N.-G., Almgren, A., Sandberg, A.-S., and Granfeldt, Y. (2015). Phytate, zinc, iron and calcium content of common Bolivian food, and implications for mineral bioavailability. Journal of Food Composition and Analysis 39, 111-119. Liang, J., Han, B.-Z., Nout, M. J. R., and Hamer, R. J. (2008). Effects of soaking, germination and fermentation on phytic acid, total and in vitro soluble zinc in brown rice. Food Chemistry 110, 821-828. Magaia, T., Uamusse, A., Sjӧholm, I., and Skog, K. (2013). Dietary fiber, organic acids and minerals in selected wild edible fruits of Mozambique. SpringerPlus 2, 88. Mahgoub, S. E. O., and Elhag, S. A. (1998). Effect of milling, soaking, malting, heat- treatment and fermentation on phytate level of four Sudanese sorghum cultivars. Food Chemistry 61(1/2), 77-80. Mangue, P. D., and Oreste, M. N. (1999). Country brief on non-wood forest products statistics- Mozambique. Data collection and analysis for sustainable forest management in ACP countries-Linking National and International EC-FAO

9

Partnership programme (1998-2000). Tropical forestry budget line B7-6201/97- 15/VIII/Fot project GCP/INT/679/EC, Mozambique. http://www.fao.org/3/a- x6693e.pdf, accessed at 19/03/2015. Mitchikpe, E. C. S., Dossa, R. A. M., Ategbo, E.-A. D., Van Raaij, J. M. A., Hulshof, P. J. M., and Kok, F. J. (2008). The supply of bioavailable iron and zinc may be affected by phytate in Beninese children. Journal of Food Composition and Analysis 21, 17-25. Muhammad, S., Hassan, L. G., Dangoggo, S. M., Hassan, S. W., Umar, K. J., and Aliyu, R. U. (2011). Nutritional and antinutritional composition of Sclerocarya birrea seed kernel Studia Universitatis “Vasile Goldiş”, Seria Ştiinţele Vieţii 21 (4), 693-699. www.studiauniversitatis.ro, accessed at 24/02/2015. Nkafamiya, I. I., Osemeahon, S. A., Dahiru, D., and Umaru, H. A. (2007). Studies on the chemical composition and physicochemical properties of the seeds of baobab (Adansonia digitata). African Journal of Biotechnology 6 (6), 756-759. Nnam, N. M., and Obiakor, P. N. (2003). Effect of fermentation on the nutrient and antinutrient composition of baobab (Adansonia digitata) seeds and rice (Oryza sativa) grains. Ecology of Food and Nutrition 42, 265-277. Osman, M. A. (2004). Chemical and nutrient analysis of baobab (Adansonia digitata) fruit and seed protein solubility. Plant Food of Human Nutrition 59, 29-33. Oyeleke, G. O., Salam, M. A., and Aderoro, R. O. (2012). Some aspects of nutrients analysis of seed, pulp and oil of baobab (Andasonia digitata L.). Journal of Environmental Science, Toxicology and Food Technology 1 (4), 32-35. Proll, J., Petzke, K. J., Ezeagu, I. E., and Metges, C. C. (1998). Low nutritional quality of unconventional tropical crop seeds in rats - Nutrient requirements and interactions. Journal of Nutrition 128, 2014-2022. Reale, A., Konietzny, U., Coppola, R., Sorrentino, E., and Greiner, R. (2007). The importance of lactic acid bacteria for phytate degradation during cereal dough fermentation. Journal of Agricultural and Food Chemistry 55, 2993-2997. Sandberg, A. S., and Svanberg, U. (1991). Phytate hydrolysis by phytase in cereals; effects on in vitro estimation of iron availability. Journal of Food Science 56 (5), 1330-1333. Saulawa, L. A., Yaradua, A. I., and Shuaibu, L. (2014). Effect of different processing methods on proximate, mineral and anti-nutritional factors content of baobab (Adansonia digitata) seeds. Pakistan Journal of Nutrition 13 (6), 314-318. Sharma, A., and Sehgal, S. (1992). Effect of processing and cooking on the antinutritional factors of faba bean (Vicia faba). Food Chemistry 43, 383-385. Sreeramulu, G., Srinivasa, D. S., Nand, K., and Joseph, R. (1996). Lactobacillus amylovorus as a phytase producer in submerged culture. Letters in Applied Microbiology 23, 385- 388. Wairagu, N. W., Kiptoo, J., and Githiomi, J. K. (2013). Nutritional assessment of Sclerocarya birrea (Amarula) fruit from Kenya International Journal of Current Research 5 (5), 1074-1078.

10 10

Table 1. Phytic acid in A. digitata and S. birrea kernels pre-treated by boiling or autoclaving for 15 minures before incubation with L. plantarum for 48 hours at 37°C, (n=2). Sample Addition of Phytic acid L. plantarum (g/100 g dry matter) A. digitata S. birrea

Boiled No 4.4 ± 0.6 2.2 ± 0.1

Boiled Yes 4.2 ± 0.0 1.9 ± 0.2

Autoclaved No 3.6 ± 0.1 1.8 ± 0.0

Autoclaved Yes 3.3 ± 0.0 1.6 ± 0.0

11

Table 2. Phytic acid content in A. digitata and S. birrea kernels incubated at 55°C with phytase for different times and then boiled for 5 minutes, (n=2).

A. digitata S. birrea Phytase Incubation time Phytic acid (mg/g sample) (min) (g/100 g dry matter)

0 0 4.0 ± 0.1 2.4 ± 0.2

10 2 3.9 ± 0.2 1.9 ± 0.3

10 15 2.9 ± 0.7 2.0 ± 0.1

10 30 3.2 ± 0.4 2.1 ± 0.3

10 60 3.3 ± 0.0 2.3 ± 0.2

10 240 2.5 ± 0.2 1.0 ± 0.1

50 601 1.9 ± 0.1 0.8 ± 0.2

150mg/ml of phytase

12 12

Table 3. Contents of iron, magnesium and zinc in the supernatant from A. digitata and S. birrea kernels incubated at 55°C with or without phytase for different times, (n=2). Phytase Incubation time Fe Mg Zn (mg/g sample) (min)/Sample (mg/100 g dry matter)

A. digitata 4.0-6.0 * 626 - 706 * 5.2-5.7 *

0 0 0.9 ± 0.09 109 ± 5.51 4.5 ± 0.04

10 2 4.4 ± 0.14 329 ± 5.65 1.3 ± 0.00

10 15 3.0 ± 0.28 404 ± 23.9 1.1 ± 0.08

10 30 2.6 ± 0.16 373 ± 21.9 0.9 ± 0.00

10 60 2.1 ± 0.12 359 ± 17.1 1.2 ± 0.01

10 240 2.1± 0.08 409 ± 24.6 1.4 ± 0.00

S. birrea 4.0* 346 - 436* 4.5*

0 0 0.2 ± 0.03 49 ± 0.03 0.7 ± 0.00

10 2 2.3 ± 0.10 299 ± 20.2 2.6 ± 0.00

10 15 2.4 ± 0.11 289 ± 17.3 2.8 ± 0.02

10 30 2.4 ± 0.09 311 ± 20.0 2.7 ± 0.01

10 60 2.1 ± 0.09 310 ± 16.2 2.2 ± 0.02

10 240 2.1 ± 0.11 344 ± 21.8 2.4 ± 0.02

*Results from a previous study on mineral content in the kernels (Magaia et al., 2013).

13

Table 4. Literature data on the phytic acid content of the seeds of A. digitata and kernels of S. birrea.

Phytic acid (%) Based on Reference

A. digitata

6.66 and 7.13 -1 Adubiaro et al., 2011

4.903* DW2 Mitchikpe et al., 2008

1.75 and 0.62 - Saulawa et al., 2014

1.40* - Proll et al., 1998

1.20 - Ezeagu, 2005

0.20 - Nkafamyia et al., 2007

0.0730* FW3 Osman, 2004

0.18 and 0.164 - Nnam & Obiakor, 2003

S. birrea

0.423* DW Muhammad et al., 2011

*= recalculated 1- = not given; 2DW= dry weight. 3FW = fresh weight; 4 = units not stated.

14 14

AU

Minutes

Figure 1: Chromatogram from the analysis of A. digitata kernels after phytase treatment. The retention time for the peak corresponding to inositol hexaphosphate is around 5 minutes.

15